Prompt word generation method and device applied to large model and electronic equipment

Abstract

The invention provides a cue word generation method and device applied to a large model and electronic equipment, relates to the field of artificial intelligence, and can be used for application scenes such as generative search, intelligent document editing, intelligent assistants, virtual assistants and intelligent e-commerce. According to the specific implementation scheme, first task information is displayed on a first interaction interface; wherein the first task information comprises related information of a large model processing task corresponding to the cue word to be generated; receiving a plurality of first candidate prompt words input by a plurality of first target objects for the first task information; and determining a target cue word in the plurality of first candidate cue words. According to the embodiment of the invention, the cue word generation efficiency of the large model can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence, in particular to the field of large models, and can be used in application scenarios such as generative search, intelligent document editing, intelligent assistants, virtual assistants, and intelligent e-commerce. Background Art

[0002] With the development of artificial intelligence technology, large models are being used more and more widely. Since large models themselves do not have the business capabilities of various application scenarios, when large models are applied to various tasks, in order for large models to complete tasks, prompt words are needed to guide the output of large models. However, the efficiency of generating prompt words is currently low and the cost is high. Summary of the invention

[0003] The present disclosure provides a method, device and electronic device for generating prompt words applied to a large model.

[0004] According to one aspect of the present disclosure, a method for generating prompt words applied to a large model is provided, comprising:

[0005] Displaying first task information on the first interactive interface; wherein the first task information includes relevant information of the large model processing task corresponding to the prompt word to be generated;

[0006] receiving a plurality of first candidate prompt words input by a plurality of first target objects with respect to the first task information;

[0007] A target cue word is determined from a plurality of first candidate cue words.

[0008] According to another aspect of the present disclosure, there is provided a device for generating prompt words applied to a large model, comprising:

[0009] A first task display module, used to display first task information on a first interactive interface; wherein the first task information includes relevant information of a large model processing task corresponding to a prompt word to be generated;

[0010] A prompt word receiving module, used for receiving a plurality of first candidate prompt words input by a plurality of first target objects with respect to the first task information;

[0011] The prompt word determination module is used to determine a target prompt word from a plurality of first candidate prompt words.

[0012] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute any method in the embodiments of the present disclosure.

[0016] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute any method according to the embodiments of the present disclosure.

[0017] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which implements any method according to the embodiments of the present disclosure when executed by a processor.

[0018] The technical solution of the disclosed embodiment can determine the target prompt word by using multiple first candidate prompt words input by multiple first target objects for the first task information, thereby improving the prompt word generation efficiency and reducing the prompt word generation cost.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0021] Figure 1 It is a flowchart of a method for generating prompt words applied to a large model provided by an embodiment of the present disclosure;

[0022] Figure 2 It is a schematic diagram of the overall process of a method for generating prompt words applied to a large model in an application example of the present disclosure;

[0023] Figure 3 is a schematic block diagram of a device for generating prompt words applied to a large model provided by an embodiment of the present disclosure;

[0024] Figure 4 is a schematic block diagram of a device for generating prompt words applied to a large model provided by another embodiment of the present disclosure;

[0025] Figure 5 is a schematic block diagram of a device for generating prompt words applied to a large model provided by another embodiment of the present disclosure;

[0026] Figure 6 is a schematic block diagram of a device for generating prompt words applied to a large model provided by another embodiment of the present disclosure;

[0027] Figure 7The block diagram of an electronic device used to implement the method for generating prompt words applied to a large model according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0028] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0029] Figure 1 A method for generating prompt words for a large model provided by an embodiment of the present disclosure is shown. The method can be applied to a device for generating prompt words for a large model, and the device can be deployed in an electronic device. The electronic device is, for example, a single or multi-machine terminal, server or other processing device. Among them, the terminal can be a user equipment (User Equipment, UE) such as a mobile device, a personal digital assistant (Personal Digital Assistant, PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc.; the server can be a stand-alone server or a server cluster. In some possible implementations, the method can also be implemented by a processor calling computer-readable instructions stored in a memory. For example Figure 1 As shown, the method may include the following steps S110 to S130.

[0030] Step S110: displaying first task information on the first interactive interface; wherein the first task information includes relevant information of the large model processing task corresponding to the prompt word to be generated.

[0031] Optionally, the first task information can be edited by a user who needs to generate prompt words, and the prompt words can be used to guide the large model to process the task so that the large model can accurately output information. Optionally, the first task information can include one or more of the following information: a background introduction of the task, a field for which prompt words are to be generated, and a usage scenario for which prompt words are to be generated.

[0032] Optionally, the first interactive interface can be used for the electronic device to interact with the first target object. The first target object can be a user. In the disclosed embodiment, the input of multiple users is obtained by displaying the first task information, so as to improve the efficiency of generating prompt words through the crowdsourcing mode. Therefore, the first target object can be understood as a crowdsourcing user. The crowdsourcing user can view the first task information on the first interactive interface. It can be understood that the crowdsourcing users can come from all walks of life and have different backgrounds and professional skills.

[0033] Optionally, users who have a need to generate prompt words can edit the first task information on the prompt word task publishing system, including the background introduction of the task, the field of prompt words to be generated, the usage scenario of the prompt words to be generated, etc., and then issue the task to the prompt word task collection and evaluation system. Optionally, the first task information can also include reward information to encourage crowdsourcing users to participate in the creation of prompt words. The reward information can include the amount of valuable information, such as currency, coupons, points, etc. The first task information can be displayed to the crowdsourcing users in the interactive interface (i.e., the first interactive interface) of the prompt word task collection and evaluation system.

[0034] Exemplary, the prompt word task publishing system and the prompt word task collection and evaluation system can be different interfaces in the same software, or different software or different modules of the same system. Users with prompt word generation requirements can log in the prompt word task publishing system, and crowdsourcing users can log in the prompt word task collection and evaluation system. Of course, the prompt word task publishing system and the prompt word task collection and evaluation system can also be different software, respectively for users with prompt word generation requirements and crowdsourcing users.

[0035] Step S120: receiving a plurality of first candidate prompt words input by a plurality of first target objects with respect to the first task information.

[0036] In the disclosed embodiment, the first target object may be understood as a crowdsourcing user. The crowdsourcing user may answer the first task information after viewing the first task information. The first candidate prompt word may be the answer content input by the crowdsourcing user for the first task information.

[0037] Optionally, the first target object may be a crowdsourcing user in a field related to the first task information, and the accuracy of the output information of the large model may be improved by the matching degree between the first candidate prompt word and the first task information. For example, when the first task information involves the communication field, multiple first target objects may be crowdsourcing users in the communication field.

[0038] In actual applications, crowdsourcing personnel can log in to the prompt word task collection and evaluation system, select a task on the prompt word task collection and evaluation system, and then give a corresponding answer (ie, the first candidate prompt word) and send the first candidate prompt word.

[0039] Step S130: determining a target prompt word from a plurality of first candidate prompt words.

[0040] In the disclosed embodiment, the target prompt word may be one or more prompt words among the plurality of first candidate prompt words that have the highest matching degree with the first task information. The target prompt word may be input into the large model to guide the large model to process the task.

[0041] In one example, a plurality of first candidate prompt words may be sent to a user who publishes the first task information, and the user who publishes the first task information determines a target prompt word from the plurality of first candidate prompt words.

[0042] In the related art, the generation of prompt words for large models mainly relies on prompt word experts to generate and judge. Since the number of prompt word experts involved is small, this method has low efficiency in generating prompt words and high labor costs. According to the technical solution of the embodiment of the present disclosure, multiple first candidate prompt words input by multiple first target objects for the first task information can be used to determine the target prompt word, which can improve the efficiency of prompt word generation and reduce the cost of prompt word generation.

[0043] In some embodiments, step S130: determining a target prompt word from a plurality of first candidate prompt words may include:

[0044] Acquiring, through the second interactive interface, a plurality of evaluation information corresponding to each of the plurality of first candidate prompt words;

[0045] Based on a plurality of evaluation information corresponding to each first candidate prompt word, a target prompt word is determined from a plurality of first candidate prompt words.

[0046] In the disclosed embodiment, each first candidate prompt word can be sent to the second interactive interface, and the second interactive interface is used to display each first candidate prompt word and the first task information. Any user can view the second interactive interface and publish evaluation information on the second interactive interface, and the evaluation information is used to indicate the degree of matching between the first candidate prompt word and the first task information. Optionally, the evaluation information may include any user's rating, comment, vote, etc. on the first candidate prompt word.

[0047] In one embodiment, the user who publishes the first task information can view multiple evaluation information corresponding to each first candidate prompt word, so as to determine the target prompt word from the multiple first candidate prompt words.

[0048] Optionally, multiple evaluation information corresponding to each first candidate prompt word can be obtained in the prompt word task collection and evaluation system, and the evaluated first candidate prompt words and the multiple evaluation information corresponding to the first candidate prompt words can be sent to the prompt vocabulary general system. The prompt vocabulary general system can determine the target prompt word based on the received multiple evaluation information corresponding to each evaluated first candidate prompt word, and submit the target prompt word to the prompt word task publishing system.

[0049] According to the above embodiment, by using the multiple evaluation information corresponding to each first candidate prompt word, the target prompt word is determined from multiple first candidate prompt words, so that the quality of the target prompt word can be improved, thereby accurately guiding the large model processing task and improving the accuracy of the large model output information.

[0050] In some embodiments, obtaining multiple evaluation information corresponding to each of the multiple first candidate prompt words through the second interactive interface may include:

[0051] For each first candidate prompt word, display the second task information corresponding to the first candidate prompt word on the second interactive interface; wherein the second task information includes relevant information of the first task information and the first candidate prompt word;

[0052] A plurality of evaluation information input by a plurality of second target objects with respect to the second task information is received.

[0053] In the disclosed embodiment, after receiving the first candidate prompt word, second task information corresponding to the first candidate prompt word may be generated, and the second task information is used to display relevant information of the first task information and the first candidate prompt word to each second target object.

[0054] In the disclosed embodiment, the second target object can be understood as a crowdsourcing user. The crowdsourcing user can select one or more second task information from the second task information corresponding to each first candidate prompt word on the second interactive interface and enter corresponding evaluation information, that is, the crowdsourcing user can evaluate the degree of match between the first candidate prompt word and the first task information.

[0055] In the disclosed embodiment, each second target object may input different evaluation information for the same second task information, and the evaluation information may include the second target object's rating, comment, vote, etc. on the first candidate prompt word.

[0056] Optionally, the evaluation information may include the evaluation of the first candidate prompt word by the second target object from multiple dimensions to improve the comprehensiveness and accuracy of the evaluation information. For example, the evaluation information may include the professionalism, clarity, etc. of the first candidate prompt word.

[0057] Optionally, the crowdsourcing personnel can log in to the prompt word task collection and evaluation system, and choose to complete the second task on the prompt word task collection and evaluation system to evaluate the quality of the first candidate prompt word. Finally, the evaluated first candidate prompt word and the evaluation information corresponding to the first candidate prompt word will be sent to the prompt vocabulary system.

[0058] According to the above embodiment, the relevant information of the first task information and the first candidate prompt word can be displayed to the second target object through the second task information, and the second target object can evaluate the quality of the first candidate prompt word, which can improve the accuracy of the evaluation information, thereby accurately determining the target prompt word from multiple first candidate prompt words.

[0059] In some embodiments, determining a target prompt word from among a plurality of first candidate prompt words based on a plurality of evaluation information corresponding to each first candidate prompt word may include:

[0060] Determine a first score corresponding to each first candidate prompt word based on a plurality of evaluation information corresponding to each first candidate prompt word;

[0061] Based on the first score corresponding to each first candidate prompt word, a target prompt word is determined from the plurality of first candidate prompt words.

[0062] In the disclosed embodiment, the evaluation information may be a score of the second target object on the first candidate prompt word, and the first preset algorithm may be used to process the multiple evaluation information corresponding to each first candidate prompt word to obtain the first score corresponding to each first candidate prompt word. Similarly, the second preset algorithm may be used to process the first score corresponding to each first candidate prompt word to determine the target prompt word from the multiple first candidate prompt words.

[0063] The following is an example of using a preset algorithm to obtain the first score:

[0064] Example 1: A weighted average algorithm can be used to analyze the objectivity of each evaluation information and the publisher of each evaluation information, and different weights can be assigned according to different evaluation information to calculate the comprehensive score of each first candidate prompt word, that is, the first score corresponding to the first candidate prompt word. Of course, the first score can also be determined by directly calculating the average of multiple evaluation information.

[0065] Example 2: Each piece of evaluation information may be sorted to obtain a sorting position of each piece of evaluation information, and the evaluation information at a preset sorting position is used as the first score.

[0066] The following is an example of using a preset algorithm to determine a target prompt word from a plurality of first candidate prompt words. The following example can be combined with the above example at will:

[0067] Example 1: The first scores corresponding to each first candidate prompt word may be ranked, and one or more first candidate prompt words with the highest rankings may be selected from multiple first candidate prompt words as target prompt words.

[0068] Example 2: When a first score corresponding to the first candidate prompt word is higher than a preset threshold, the first candidate prompt word may be determined as the target prompt word.

[0069] According to the above embodiment, through the first score corresponding to each first candidate prompt word, the target prompt word can be accurately determined from multiple first candidate prompt words, and the quality of the target prompt word can be further improved, thereby guiding the large model to process the task more accurately, and further improving the accuracy of the output information of the large model.

[0070] In some embodiments, the first task information includes a task deadline;

[0071] The method for generating prompt words applied to the large model may further include:

[0072] In response to obtaining any evaluation information corresponding to a first candidate prompt word among the plurality of first candidate prompt words, storing the evaluation information in a storage area;

[0073] In response to reaching the task deadline, all evaluation information corresponding to the first candidate prompt words is extracted from the storage area to determine a target prompt word from a plurality of first candidate prompt words.

[0074] In the disclosed embodiment, before the task deadline, each evaluation information of each first candidate prompt word can be obtained in real time and stored in the storage area. When the task deadline is reached, all first candidate prompt words and all corresponding evaluations can be extracted from the storage area.

[0075] Exemplarily, the storage area may be a local storage area of ​​the prompt word generating device, or may be a cloud storage space.

[0076] According to the above embodiment, the evaluation information can be acquired and stored before the task deadline, so as to ensure that the evaluation information corresponding to the first candidate prompt word is not omitted in the subsequent process of determining the target prompt word.

[0077] In one embodiment, in response to reaching a preset time threshold, all evaluation information corresponding to the first candidate prompt word within the time period can be extracted from the storage area, that is, the evaluation information corresponding to the first candidate prompt word is regularly extracted from the storage area at regular intervals. Compared with extracting all the evaluation information at one time, the pressure on the storage area can be reduced and the efficiency of extracting the evaluation information can be improved.

[0078] In some embodiments, the first task information is released by a third target object; the above step S130: determining a target prompt word from a plurality of first candidate prompt words includes:

[0079] Determine one or more second candidate prompt words from among the plurality of first candidate prompt words;

[0080] Displaying one or more second candidate prompt words to a third target object through a third interactive interface;

[0081] A target prompt word is determined based on a selection operation of the third target object among one or more second candidate prompt words.

[0082] In the embodiment of the present disclosure, a preset algorithm can be used to determine one or more second candidate prompt words from multiple first candidate prompt words. For example, one or more second candidate prompt words can be determined from multiple first candidate prompt words based on multiple evaluation information corresponding to each first candidate prompt word, and then the target prompt word selected by the third target object is obtained through the third interactive interface. The specific preset algorithm can refer to the relevant description of the corresponding steps in any of the above embodiments, and will not be repeated here.

[0083] It can be understood that, when only one second candidate prompt word is displayed, the third target object can choose to use the second candidate prompt word, that is, the second candidate prompt word is the target prompt word. The third target object can also choose not to use the second candidate prompt word, in which case other candidate prompt words generated by the algorithm can be displayed to the third target object through the third interactive interface; or the third target object can extend the task deadline, and when the task deadline is reached, the new second candidate prompt word is obtained by returning to execute the above step S120, so as to display the new second candidate prompt word to the third target object.

[0084] It can be understood that, when multiple second candidate prompt words are displayed, the third target object can select from the multiple second candidate prompt words and use the selected one or more second candidate prompt words as the target prompt word.

[0085] Optionally, one or more second candidate prompt words can be submitted to the prompt word task publishing system, and the third target object can select the second candidate prompt word to be adopted in the prompt word task publishing system. Optionally, the valuable information corresponding to the reward information in the first task information can also be provided to the first target object who inputs the second candidate prompt word.

[0086] According to the above embodiment, one or more second candidate prompt words may be displayed to the third target object so that the third target object can make a selection, thereby improving the usage experience of the third target object and improving the quality of the target prompt words.

[0087] In order to facilitate understanding of the above technical solution, Figure 2The overall flow diagram of the method for generating prompt words applied to a large model in an application example of the present disclosure is provided. The third target object can edit the first task information in the prompt word task publishing system 20, and send the first task information to the collection and evaluation software 21 and the collection and evaluation software 22. The first target object and the second target object can respectively input answers (i.e., first candidate prompt words) in the collection and evaluation software 21 and the collection and evaluation software 22, and the first target object and the second target object can respectively evaluate the answers input by others. The collection and evaluation software 21 and the collection and evaluation software 22 can respectively submit the answers and the evaluation information corresponding to the answers to the prompt word task summary system 23. The prompt word task summary system 23 can screen the answers and submit the screened answers (i.e., second candidate prompt words) to the task publishing system 20, and the third target object can determine the target prompt word from the second candidate prompt words.

[0088] According to an embodiment of the present disclosure, the present disclosure also provides a device for generating prompt words applied to a large model, Figure 3 A schematic block diagram of a device for generating prompt words applied to a large model provided by an embodiment of the present disclosure is shown. Figure 3 As shown, the device comprises:

[0089] The first task display module 310 is used to display the first task information on the first interactive interface; wherein the first task information includes relevant information of the large model processing task corresponding to the prompt word to be generated;

[0090] The prompt word receiving module 320 is used to receive a plurality of first candidate prompt words input by a plurality of first target objects with respect to the first task information;

[0091] The prompt word determination module 330 is used to determine a target prompt word from a plurality of first candidate prompt words.

[0092] In some embodiments, Figure 4 As shown, the prompt word determination module 330 may include:

[0093] The evaluation acquisition submodule 410 is used to acquire multiple evaluation information corresponding to each of the multiple first candidate prompt words through the second interactive interface;

[0094] The prompt word determination submodule 420 is used to determine a target prompt word from a plurality of first candidate prompt words based on a plurality of evaluation information corresponding to each first candidate prompt word.

[0095] In some embodiments, Figure 5 As shown, the evaluation acquisition submodule 410 may include:

[0096] The second task display unit 510 is used to display, for each first candidate prompt word, the second task information corresponding to the first candidate prompt word on the second interactive interface; wherein the second task information includes the relevant information of the first task information and the first candidate prompt word;

[0097] The evaluation receiving unit 520 is configured to receive a plurality of evaluation information input by a plurality of second target objects with respect to the second task information.

[0098] In some embodiments, the prompt word determination submodule 420 is specifically used to:

[0099] Determine a first score corresponding to each first candidate prompt word based on a plurality of evaluation information corresponding to each first candidate prompt word;

[0100] Based on the first score corresponding to each first candidate prompt word, a target prompt word is determined from the plurality of first candidate prompt words.

[0101] In some embodiments, the first task information includes a task deadline;

[0102] like Figure 6 As shown, the generating device for the prompt word applied to the large model further includes a storage module 610, and the storage module 610 is used to:

[0103] In response to obtaining any evaluation information corresponding to a first candidate prompt word among the plurality of first candidate prompt words, storing the evaluation information in a storage area;

[0104] In response to reaching the task deadline, all evaluation information corresponding to the first candidate prompt words is extracted from the storage area to determine a target prompt word from a plurality of first candidate prompt words.

[0105] In some embodiments, the first task information is published by a third target object;

[0106] The prompt word determination module 330 is specifically used for:

[0107] Determine one or more second candidate prompt words from among the plurality of first candidate prompt words;

[0108] Displaying one or more second candidate prompt words to a third target object through a third interactive interface;

[0109] A target prompt word is determined based on a selection operation of the third target object among one or more second candidate prompt words.

[0110] For the description of specific functions and examples of each module and submodule of the device in the embodiment of the present disclosure, reference can be made to the relevant description of the corresponding steps in the above method embodiment, which will not be repeated here.

[0111] In the technical solution disclosed herein, the acquisition, storage and application of the target object's personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0112] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0113] Figure 7 A schematic block diagram of an example electronic device 700 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0114] like Figure 7 As shown, the device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 707 to a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 804. An input / output (I / O) interface 705 is also connected to the bus 704.

[0115] A number of components in the device 700 are connected to the I / O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0116] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above, such as the generation method of the prompt word applied to the large model. For example, in some embodiments, the generation method of the prompt word applied to the large model may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 700 via the ROM 702 and / or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the generation method of the prompt word applied to the large model described above may be executed. Alternatively, in other embodiments, the computing unit 701 may be configured in any other appropriate manner (for example, by means of firmware) to execute the method for generating prompt words applied to the large model.

[0117] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0118] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0119] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0120] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0121] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

[0122] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0123] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0124] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for generating prompt words applied to a large model, comprising: Displaying first task information on the first interactive interface; wherein the first task information includes relevant information of the large model processing task corresponding to the prompt word to be generated; receiving a plurality of first candidate prompt words input by a plurality of first target objects with respect to the first task information; A target prompt word is determined from the plurality of first candidate prompt words.

2. The method according to claim 1, wherein: The step of determining a target prompt word from among the plurality of first candidate prompt words comprises: Acquiring, through the second interactive interface, a plurality of evaluation information corresponding to each of the plurality of first candidate prompt words; Based on a plurality of evaluation information corresponding to each of the first candidate prompt words, a target prompt word is determined from the plurality of first candidate prompt words.

3. The method according to claim 2, wherein: The obtaining, through the second interactive interface, a plurality of evaluation information corresponding to each of the plurality of first candidate prompt words includes: For each of the first candidate prompt words, displaying the second task information corresponding to the first candidate prompt word on the second interactive interface; wherein the second task information includes relevant information of the first task information and the first candidate prompt word; A plurality of evaluation information input by a plurality of second target objects with respect to the second task information is received.

4. The method according to claim 2 or 3, wherein: The step of determining the target prompt word from among the plurality of first candidate prompt words based on the plurality of evaluation information corresponding to each first candidate prompt word comprises: Determining a first score corresponding to each first candidate prompt word based on a plurality of evaluation information corresponding to each first candidate prompt word; Based on the first score corresponding to each of the first candidate prompt words, a target prompt word is determined from the plurality of first candidate prompt words.

5. The method according to any one of claims 2 to 4, wherein: The first task information includes a task deadline; The method further comprises: In response to acquiring any evaluation information corresponding to a first candidate prompt word among the plurality of first candidate prompt words, storing the evaluation information in a storage area; In response to reaching the task deadline, all evaluation information corresponding to the first candidate prompt words is extracted from the storage area to determine a target prompt word from among the plurality of first candidate prompt words.

6. The method according to any one of claims 1 to 5, wherein: The first task information is published by a third target object; The step of determining a target prompt word from among the plurality of first candidate prompt words comprises: Determine one or more second candidate prompt words from the plurality of first candidate prompt words; Displaying the one or more second candidate prompt words to the third target object through a third interactive interface; The target prompt word is determined based on a selection operation of the third target object among the one or more second candidate prompt words.

7. A device for generating prompt words for a large model, comprising: A first task display module, used to display first task information on a first interactive interface; wherein the first task information includes relevant information of a large model processing task corresponding to a prompt word to be generated; A prompt word receiving module, used for receiving a plurality of first candidate prompt words input by a plurality of first target objects with respect to the first task information; The prompt word determination module is used to determine a target prompt word from the plurality of first candidate prompt words.

8. The device according to claim 7, wherein: The prompt word determination module includes: An evaluation acquisition submodule, used for acquiring a plurality of evaluation information corresponding to each of the plurality of first candidate prompt words through a second interactive interface; The prompt word determination submodule is used to determine a target prompt word from the plurality of first candidate prompt words based on a plurality of evaluation information corresponding to each of the first candidate prompt words.

9. The device according to claim 8, wherein: The evaluation acquisition submodule includes: A second task display unit, configured to display, for each of the first candidate prompt words, second task information corresponding to the first candidate prompt word on the second interactive interface; wherein the second task information includes relevant information of the first task information and the first candidate prompt word; The evaluation receiving unit is used to receive a plurality of evaluation information input by a plurality of second target objects with respect to the second task information.

10. The device according to claim 8 or 9, wherein: The prompt word determination submodule is specifically used for: Determining a first score corresponding to each first candidate prompt word based on a plurality of evaluation information corresponding to each first candidate prompt word; Based on the first score corresponding to each of the first candidate prompt words, a target prompt word is determined from the plurality of first candidate prompt words.

11. The device according to any one of claims 8 to 10, wherein: The first task information includes a task deadline; The device further comprises a storage module, wherein the storage module is used for: In response to acquiring any evaluation information corresponding to a first candidate prompt word among the plurality of first candidate prompt words, storing the evaluation information in a storage area; In response to reaching the task deadline, all evaluation information corresponding to the first candidate prompt words is extracted from the storage area to determine a target prompt word from among the plurality of first candidate prompt words.

12. The device according to any one of claims 7 to 11, wherein: The first task information is published by a third target object; The prompt word determination module is specifically used for: Determine one or more second candidate prompt words from the plurality of first candidate prompt words; Displaying the one or more second candidate prompt words to the third target object through a third interactive interface; The target prompt word is determined based on a selection operation of the third target object among the one or more second candidate prompt words.

13. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-6.

15. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images