Prompt word template generation method and device, electronic equipment and storage medium

Abstract

The invention provides a cue word template generation method and device, electronic equipment and a storage medium, and relates to the technical field of computers, in particular to the technical field of artificial intelligence such as natural language processing, large language models and deep learning. According to the specific implementation scheme, a first data set and an initial cue word template corresponding to a target task are obtained, and the first data set comprises sample data, text description corresponding to the sample data and reference answers; determining a first answer corresponding to the sample data according to the initial cue word template, the sample data and the corresponding text description; determining a first evaluation result corresponding to the first answer according to the reference answer and the first answer; and based on the first evaluation result and the text description, optimizing the initial cue word template to obtain a target cue word template.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, in particular to the field of artificial intelligence technology such as natural language processing, large language models, and deep learning, and specifically to a method, device, electronic device, and storage medium for generating a prompt word template. Background Art

[0002] In recent years, Large Language Model (LLM) has gained wide attention in the field of natural language processing. Through large-scale pre-training and fine-tuning, LLM has demonstrated excellent performance in a variety of tasks. However, studies have shown that the output quality of LLM depends largely on the design of the input prompt. Even for the same task, the input of different prompts may lead to significant differences in the accuracy, fluency, and relevance of the generated results. Therefore, prompt design and optimization have become the core issues of prompt engineering. Summary of the invention

[0003] The present disclosure provides a method, device, electronic device and storage medium for generating a prompt word template.

[0004] According to a first aspect of the present disclosure, a method for generating a prompt word template is provided, comprising:

[0005] Obtaining a first data set and an initial prompt word template corresponding to the target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer;

[0006] Determine a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description;

[0007] Determine a first evaluation result corresponding to the first answer according to the reference answer and the first answer;

[0008] Based on the first evaluation result and the text description, the initial prompt word template is optimized to obtain a target prompt word template.

[0009] According to a second aspect of the present disclosure, a device for generating a prompt word template is provided, comprising:

[0010] An acquisition module, used to acquire a first data set and an initial prompt word template corresponding to a target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer;

[0011] A first determination module, configured to determine a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description;

[0012] A second determination module, configured to determine a first evaluation result corresponding to the first answer according to the reference answer and the first answer;

[0013] An optimization module is used to optimize the initial prompt word template based on the first evaluation result and the text description to obtain a target prompt word template.

[0014] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] 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 so that the at least one processor can execute the method for generating a prompt word template as described in the first aspect.

[0018] According to a fourth 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 the method for generating a prompt word template as described in the first aspect.

[0019] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising computer instructions, which, when executed by a processor, implement the steps of the method for generating a prompt word template as described in the first aspect.

[0020] The method, device, electronic device and storage medium for generating a prompt word template provided by the present disclosure have the following beneficial effects:

[0021] In the disclosed embodiment, the first answer corresponding to the sample data is determined according to the initial prompt word template, the sample data and the corresponding text description, and then the first evaluation result corresponding to the first answer is determined according to the reference answer corresponding to the sample data and the first answer, and finally the initial prompt word template is optimized based on the first evaluation result and the text description to obtain the target prompt word template. Thus, the defects of the initial prompt word template can be determined by the difference between the first answer corresponding to the sample data under the initial prompt word template and the reference answer, and then the initial prompt word template is optimized in combination with the text description corresponding to the sample data, thereby realizing the automatic optimization of the prompt word template, improving the efficiency and accuracy of the optimization of the prompt word template, and thus improving the quality of the optimized target prompt word template.

[0022] 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

[0023] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0024] Figure 1 It is a flowchart of a method for generating a prompt word template provided according to an embodiment of the present disclosure;

[0025] Figure 2 is a flowchart of a method for generating a prompt word template provided according to another embodiment of the present disclosure;

[0026] Figure 3 is a flowchart of a method for generating a prompt word template provided according to another embodiment of the present disclosure;

[0027] Figure 4 It is a structural schematic diagram of a device for generating a prompt word template according to an embodiment of the present disclosure;

[0028] Figure 5 The block diagram is a block diagram of an electronic device for implementing the method for generating a prompt word template according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] 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 and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0030] The embodiments of the present disclosure relate to artificial intelligence technology fields such as large language models, natural language processing, and deep learning.

[0031] Artificial Intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence.

[0032] Large Language Model (LLM), or large model, is a deep learning model trained based on massive text data. It can not only generate natural language text, but also deeply understand the meaning of the text and handle various natural language tasks such as text summarization, question answering, translation, etc.

[0033] Natural Language Processing (NLP) is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that can achieve effective communication between people and computers using natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Natural language processing is not a general study of natural language, but the development of computer systems that can effectively realize natural language communication, especially the software systems therein.

[0034] Deep learning is the process of learning the inherent laws and representation levels of sample data. The information obtained in the learning process is of great help in interpreting data such as text, images, and sounds. The ultimate goal of deep learning is to enable machines to have analytical learning capabilities like humans and to be able to recognize data such as text, images, and sounds.

[0035] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0036] The following describes a method, device, electronic device, and storage medium for generating a prompt word template according to an embodiment of the present disclosure with reference to the accompanying drawings.

[0037] It should be noted that the executor of the prompt word template generation method of this embodiment is a prompt word template generation device, which can be implemented by software and / or hardware, and can be configured in an electronic device, which may include but is not limited to a terminal, a server, etc.

[0038] Figure 1 It is a flowchart of a method for generating a prompt word template provided according to an embodiment of the present disclosure.

[0039] like Figure 1 As shown, the method for generating the prompt word template includes:

[0040] S101, obtaining a first data set and an initial prompt word template corresponding to a target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer.

[0041] The target task may be an image recognition task, an image description task, a text summarization task, etc. This disclosure does not limit this.

[0042] The initial prompt word template is a prompt word template generated in advance according to the target task and to be further optimized.

[0043] In some embodiments, the sample data may be at least one of the following: text data, image data, and audio data. Thus, a prompt word template corresponding to the target task can be generated for multimodal data, thereby improving the versatility of the method for generating prompt word templates.

[0044] In some embodiments, the text description is text information associated with the sample data. For example, when the target task is to describe an image and the sample data is an image, the corresponding text description may be a theme.

[0045] In some embodiments, the reference answer may be the answer corresponding to the sample data under the target task. For example, if the target task is image description and the sample data is an image, the corresponding reference answer may be the description corresponding to the image determined manually.

[0046] In some embodiments, the first data set may include one or more pieces of sample data, which is not limited in the present disclosure.

[0047] S102, determining a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description.

[0048] In some embodiments, the text description can be filled into a preset position in the initial prompt word template to obtain a first prompt word corresponding to the sample data, and then the first prompt word and the sample data are input into the first large model to obtain a first answer corresponding to the sample data. In this way, the first answer corresponding to the sample data under the initial prompt word template can be quickly obtained.

[0049] The first answer may be a first answer corresponding to the sample data generated by the large model based on the first prompt word and the sample data.

[0050] In some embodiments, the first large model may also be referred to as a generator.

[0051] In some embodiments, the first large model can be obtained by fine-tuning an existing large language model based on the target task.

[0052] S103, determining a first evaluation result corresponding to the first answer according to the reference answer and the first answer.

[0053] In some embodiments, when the target task is a preset task, the reference answer, the first answer and the second prompt word can be input into the second large model to obtain the first evaluation result, wherein the second prompt word is used to prompt the second large model to evaluate the first answer based on the reference answer. Thus, the first evaluation result corresponding to the first answer can be quickly and accurately determined based on the second large model.

[0054] The preset tasks may be tasks with complex reference answers, such as image description tasks, text summary tasks, and other tasks with long answers.

[0055] It should be noted that the text comparison task is a relatively common task, and the existing large model already has this task and can accurately compare two texts. Therefore, the existing large model with the text comparison task can be directly determined as the second large model.

[0056] In some embodiments, when the target task is not a preset task, the first answer can also be evaluated based on a simple logic evaluation function. For example, the content of the first answer is relatively simple, for example, the target task is a classification task, and the first answer and the reference answer are probabilities of belonging to each category, then the probability of the first answer and the reference answer can be directly compared based on the logic function to obtain the first evaluation result.

[0057] S104: Based on the first evaluation result and the text description, the initial prompt word template is optimized to obtain a target prompt word template.

[0058] In some embodiments, the first evaluation result, the text description, the initial prompt word template, and the third prompt word can be input into the third largest model to obtain the target prompt word template. The third prompt word is used to prompt the third largest model to optimize the initial prompt word template based on the first evaluation result and the corresponding text description.

[0059] In some embodiments, the third large model, the second large model, and the first large model may be the same large model or different large models, which is not limited in the present disclosure.

[0060] In some embodiments, the target sample data for which the first evaluation result indicates an error can be determined first, and then the first evaluation result, text description, initial prompt word template and third prompt word corresponding to the target sample data are input into the third largest model to obtain the target prompt word template. The third prompt word is used to prompt the third largest model to optimize the initial prompt word template based on the first evaluation result and the corresponding text description.

[0061] In some embodiments, if the first evaluation result includes a question that indicates a difference between the first answer and the reference answer, the first evaluation result indicates an error. It should be noted that only the first evaluation result indicating an error has reference value for optimizing the initial prompt word template. Therefore, it is necessary to first determine the target sample data for which the first evaluation result indicates an error, and then optimize the initial prompt word template based on the corresponding first evaluation result.

[0062] In some embodiments, the target sample data for which the first evaluation result indicates an error is determined, and then, when there are at least two target sample data, the text description associated with the first evaluation result is determined based on the first evaluation result and the text description corresponding to the target sample data, and then the initial prompt word template, the third prompt word, the first evaluation result, and the text description associated with the first evaluation result are input into the third large model to obtain the target prompt word template, wherein the third prompt word is used to prompt the third large model to optimize the initial prompt word template based on the first evaluation result and the corresponding text description. Thus, when there are multiple first evaluation results, the association between the first evaluation result and the text description is told to the large model, so that the target prompt word template generated by the large model can be applied to sample data corresponding to different text descriptions.

[0063] In the disclosed embodiment, the first answer corresponding to the sample data is determined according to the initial prompt word template, the sample data and the corresponding text description, and then the first evaluation result corresponding to the first answer is determined according to the reference answer corresponding to the sample data and the first answer, and finally the initial prompt word template is optimized based on the first evaluation result and the text description to obtain the target prompt word template. Thus, the defects of the initial prompt word template can be determined by the difference between the first answer corresponding to the sample data under the initial prompt word template and the reference answer, and then the initial prompt word template is optimized in combination with the text description corresponding to the sample data, thereby realizing the automatic optimization of the prompt word template, improving the efficiency and accuracy of the optimization of the prompt word template, and thus improving the quality of the optimized target prompt word template.

[0064] Figure 2 is a flowchart of a method for generating a prompt word template provided according to another embodiment of the present disclosure;

[0065] like Figure 2 As shown, the method for generating the prompt word template includes:

[0066] S201, obtaining a first data set and an initial prompt word template corresponding to a target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer.

[0067] S202: Determine a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description.

[0068] S203: Determine a first evaluation result corresponding to the first answer based on the reference answer and the first answer.

[0069] S204: Based on the first evaluation result and the text description, the initial prompt word template is optimized to obtain a target prompt word template.

[0070] The specific implementation forms of steps S201 to S204 can refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0071] S205: Determine a second answer corresponding to the sample data according to the target prompt word template, the sample data and the corresponding text description.

[0072] In some embodiments, the text description can be filled into a preset position in the target prompt word template to obtain a new prompt word corresponding to the sample data, and then the new prompt word and the sample data can be input into the first large model to obtain a second answer corresponding to the sample data.

[0073] The second answer may be the first answer corresponding to the sample data generated by the large model based on the sample data and the new prompt word corresponding to the sample data.

[0074] S206: Determine a second evaluation result corresponding to the second answer based on the second answer and the reference answer.

[0075] In some embodiments, the reference answer, the second answer and the second prompt word can be input into the second largest model to obtain a second evaluation result, wherein the second prompt word is used to prompt the second largest model to evaluate the second answer based on the reference answer.

[0076] S207: Determine a prompt word template to be optimized from the target prompt word template and the initial prompt word template.

[0077] In some embodiments, the prompt word template with the best quality can be determined from the target prompt word template and the initial prompt word template as the prompt word template to be optimized. Thus, the prompt word template with the best quality can be optimized in each iteration.

[0078] In some embodiments, a first score corresponding to the initial prompt word template and a second score corresponding to the target prompt word template may be determined, and then the prompt word template with the highest score among the first score and the second score may be determined as the prompt word template to be optimized. The higher the score, the better the quality of the prompt word template. Thus, according to the scores of each prompt word template, a prompt word template with better quality may be accurately determined as the prompt word template to be optimized.

[0079] In some embodiments, a first score corresponding to the initial prompt word template may be determined according to a first evaluation result corresponding to each sample data, and a second score corresponding to the target prompt word template may be determined according to a second evaluation result corresponding to each sample data.

[0080] In some embodiments, the third score corresponding to each first answer can be determined based on the difference between the first answer corresponding to each sample data and the reference answer, and then the first score corresponding to the initial prompt word template can be determined based on the third score corresponding to each first answer. Thus, the first score of the initial prompt word template can be comprehensively determined based on the score of the first answer corresponding to each sample data under the first prompt word template, so that the determined first score corresponding to the initial prompt word template is more accurate.

[0081] In some embodiments, the reference answer, the first answer, and the second prompt word can be output to the second largest model to obtain the first evaluation result and the first score, wherein the second prompt word is used to prompt the second largest model to evaluate the first answer based on the reference answer and output the score of the first answer. Thus, the first score can be obtained at the same time as the first evaluation result, which improves the efficiency of optimizing the initial prompt word target.

[0082] In some embodiments, a third score corresponding to the first answer may also be determined based on the similarity between the first answer and the reference answer.

[0083] In some embodiments, when there are multiple sample data, the mean, sum, variance, etc. of the third scores corresponding to the multiple first answers can be determined as the first score corresponding to the initial prompt word template.

[0084] In some embodiments, the fourth score corresponding to each second answer is determined based on the difference between the second answer corresponding to each sample data and the reference answer, and the second score corresponding to the target prompt word template is determined based on the fourth score corresponding to each second answer. Thus, the second score of the target prompt word template can be comprehensively determined based on the score of the second answer corresponding to each sample data under the target prompt word template, so that the second score corresponding to the determined target prompt word template is more accurate.

[0085] In some embodiments, the reference answer, the second answer, and the second prompt word are output to the second largest model to obtain a second evaluation result and a second score, wherein the second prompt word is used to prompt the second largest model to evaluate the second answer based on the reference answer and output the score of the second answer. Thus, the first score can be obtained at the same time as the first evaluation result, which improves the efficiency of optimizing the initial prompt word target.

[0086] In some embodiments, a third score corresponding to the first answer may also be determined based on the similarity between the first answer and the reference answer.

[0087] In some embodiments, when there are multiple sample data, the mean, sum, variance, etc. of the third scores corresponding to the multiple first answers can be determined as the first score corresponding to the initial prompt word template.

[0088] S208: Optimize the prompt word template to be optimized based on the second evaluation result and the text description to obtain an optimized target prompt word template.

[0089] In some embodiments, the second evaluation result, the text description, the prompt word template to be optimized, and the third prompt word can be input into the third largest model to obtain the optimized target prompt word template. The third prompt word is used to prompt the third largest model to optimize the prompt word template to be optimized based on the second evaluation result and the corresponding text description.

[0090] In some embodiments, after determining the optimized target prompt word template, the third evaluation result corresponding to the optimized target prompt word template can also be determined, and the prompt word template to be further optimized can be determined from the optimized target prompt word template and the prompt word template to be optimized, and based on the third evaluation result and text description, the prompt word template to be further optimized can be optimized, and this iteration is performed until the preset stop condition is reached.

[0091] The preset stop condition may be that a preset number of iterations is reached, or that the score corresponding to the optimized prompt word template reaches a preset score, or that the difference between the scores corresponding to the prompt word templates optimized after multiple consecutive iterations is less than a difference threshold.

[0092] In some embodiments, when the prompt word template to be optimized is the initial prompt word template, the first evaluation result and the second evaluation result corresponding to the sample data are merged to obtain the third evaluation result corresponding to the sample data, and then the initial prompt word is optimized based on the third evaluation result and the text description to obtain the optimized target prompt word template. In this way, the problems existing in the initial prompt word template can be more comprehensively determined, and then the initial prompt word template can be accurately optimized to improve the quality of the optimized target prompt word template.

[0093] In the disclosed embodiment, after optimizing the initial prompt word template to obtain the target prompt word template, the second answer corresponding to the sample data can be determined according to the target prompt word template, the sample data and the corresponding text description, and the second evaluation result corresponding to the second answer can be determined according to the second answer and the reference answer, and the prompt word template to be optimized can be determined from the target prompt word template and the initial prompt word template, and finally, based on the second evaluation result and the text description, the prompt word template to be optimized can be optimized to obtain the optimized target prompt word template. Thus, the prompt word template with the best quality can be optimized based on the evaluation result corresponding to the latest prompt word template, so that the prompt word template with the best quality can be optimized based on the defects that may exist in the prompt word template with the best quality, and the accuracy of the optimized prompt word template can be further improved.

[0094] Figure 3 is a flowchart of a method for generating a prompt word template provided according to another embodiment of the present disclosure;

[0095] like Figure 3 As shown, the method for generating the prompt word template includes:

[0096] S301, obtaining a first data set and an initial prompt word template corresponding to a target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer.

[0097] S302: Determine a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description.

[0098] S303: Determine a first evaluation result corresponding to the first answer based on the reference answer and the first answer.

[0099] S304, determining that the first evaluation result indicates erroneous target sample data.

[0100] The specific implementation forms of steps S301 to S304 can refer to the detailed descriptions in other embodiments of the present disclosure, and will not be described in detail here.

[0101] S305 , when there are at least two target sample data, cluster the first evaluation results corresponding to the target sample data to determine the target evaluation results corresponding to the initial prompt word template.

[0102] It should be noted that, when the number of target sample data is at least two, there may be a situation where the first evaluation results corresponding to multiple target sample data are the same, similar, or partially overlapped. Therefore, the first evaluation results can be clustered to obtain one or more different target evaluation results corresponding to the initial prompt word template, thereby reducing the input of the third largest model, thereby reducing the amount of data processed by the third largest model, and improving the optimization efficiency of the initial prompt word template.

[0103] S306: Determine a text description associated with the target evaluation result based on the target sample data to which the first evaluation result corresponding to the target evaluation result belongs and the text description corresponding to the target sample data.

[0104] In the disclosed embodiment, a target evaluation result to which the first evaluation result corresponding to the same sample data belongs and a text description corresponding to the same sample data can be associated, so that the association between the target evaluation result and the text description can be input into the third model.

[0105] S307: Input the initial prompt word template, the third prompt word, the target evaluation result, and the text description associated with the target evaluation result into the third language model to obtain the target prompt word template.

[0106] The third prompt word is used to prompt the third language model to optimize the initial prompt word template based on the target evaluation result and the associated text description.

[0107] In the disclosed embodiment, when the first evaluation result indicates that there are at least two target sample data that are wrong, the first evaluation results corresponding to the target sample data are clustered to determine the target evaluation results corresponding to the initial prompt word template, and then based on the target sample data to which the first evaluation result corresponding to the target evaluation result belongs and the text description corresponding to the target sample data, the text description associated with the target evaluation result is determined, and finally the initial prompt word template, the third prompt word, the target evaluation result, and the text description associated with the target evaluation result are input into the third language model to obtain the target prompt word template. Thus, the first evaluation results are clustered to obtain one or more different target evaluation results corresponding to the initial prompt word template, thereby reducing the input of the third model, thereby reducing the amount of data processed by the third model, and improving the efficiency of optimizing the initial prompt word template.

[0108] For example, the text description is: "Picture of black scab on the back of proliferative pemphigus", and the initial prompt word template is:

[0109] "You are a health expert who is very good at image description. You are responsible for providing a professional and objective description for a picture with the theme "{title}". Please make sure that your description is closely related to the theme of the picture, use pure simplified Chinese, and strictly avoid any special tags or formats, such as '#end#', '#END#', '#CONTENT#', '#note#', '#RESPONSE#', etc. At the same time, your description should not be left blank or produce meaningless output.

[0110] When describing images, please follow these guidelines:

[0111] *Statements about the educational value of the image or how it can help people are not allowed.

[0112] * Comments on image quality are prohibited.

[0113] *Describe according to the image type, such as inspection documents, instructions, cartoon-like popular science images, or images involving human body parts and organs, etc.

[0114] *If the image contains text, please indicate where the text is and briefly describe its content.

[0115] #OBJECTIVE OF STEP1#

[0116] Please use 2-3 keywords to succinctly and accurately summarize the core points of the image, including key people, disease symptoms, objects or indicators, etc.

[0117] Format requirements: "Brief description:"

[0118] #OBJECTIVE OF STEP2#

[0119] Based on STEP 1, further elaborate on the content of the picture, avoid duplication with the content of STEP 1, and summarize it in 1-2 sentences. When describing, please stick to the theme of the picture, do not make disease judgments or give medical advice, and do not describe too much about the symptoms or body parts that may be involved.

[0120] Format requirements: "Detailed explanation:"

[0121] #STYLE#

[0122] Please use professional medical terminology, keep the language rigorous, and make sure the sentence does not contain special marks or formats such as '#a certain word#'.

[0123] #TONE#

[0124] Express in a popular science way, the language should be easy to understand, and the sentences should be fluent and objective.

[0125] #AUDIENCE#

[0126] For people with non-medical backgrounds”

[0127] Then the first answer output by the first model can be: "Back, rash, scabs. Detailed explanation: This picture shows a person's back with obvious rashes and scabs. The rash is red and the scabs are brown, covering most of the back. The rash and scabs look dry and even have cracks in some places. This picture may be used for medical education or diagnosis to help doctors and patients understand this skin disease. However, please note that this picture cannot be used for diagnosis or treatment, because there are many types of skin diseases, which require professional diagnosis and treatment by doctors."

[0128] If the corresponding reference answer is "This picture shows the skin lesions on a person's back. The lesion area is large, almost affecting the entire back skin. Redness, swelling, ulcers and scabs can be seen. The scabs are brown and flaky, covering the surface of the skin.", then the first evaluation result output by the second largest model can be "The generated description is relatively complete, but contains redundant information, and some descriptions are not completely consistent with the full score description."

[0129] The target prompt word template can be:

[0130] "You are a health expert who is very good at image description. You are responsible for providing a professional and objective description for a picture with the theme "{title}". Please make sure that your description is closely related to the theme of the picture, use pure simplified Chinese, and strictly avoid any special tags or formats, such as '#end#', '#END#', '#CONTENT#', '#note#', '#RESPONSE#', etc. At the same time, your description should not be left blank or produce meaningless output.

[0131] When describing images, please follow these guidelines:

[0132] *Statements about the educational value of the image or how it can help people are not allowed.

[0133] * Comments on image quality are prohibited.

[0134] *Describe according to the image type, such as inspection documents, instructions, cartoon-like popular science images, or images involving human body parts and organs, etc.

[0135] *If the image contains text, please indicate where the text is and briefly describe its content.

[0136] #OBJECTIVE OF STEP1#

[0137] Please use 2-3 keywords to succinctly and accurately summarize the core points of the image, including key people, disease symptoms, objects or indicators, etc. Make sure these keywords are closely related to the theme of the picture and avoid using vague or general words.

[0138] Format requirements: "Brief description:"

[0139] #OBJECTIVE OF STEP2#

[0140] Based on STEP 1, further elaborate on the content of the picture, avoid duplication with the content of STEP 1, and summarize it in 1-2 sentences. When describing, please stick to the theme of the picture, do not make disease judgments or give medical advice, and do not describe too much about the symptoms or body parts that may be involved. Make sure the description is specific, accurate, and highly consistent with the content of the picture.

[0141] Format requirements: "Detailed explanation:"

[0142] #STYLE#

[0143] Please use professional medical terms, keep the language rigorous, and make sure that the sentence does not contain special tags or formats such as '#some word#'. Avoid using overly complex or obscure vocabulary and make sure the description is easy to understand.

[0144] #TONE#

[0145] Express in a popular science way, the language should be easy to understand, and the sentences should be fluent and objective. Avoid using highly subjective words or expressions, and ensure that the description is objective and neutral.

[0146] #AUDIENCE#

[0147] It is aimed at people with non-medical backgrounds, so please use easy-to-understand language and expressions when describing it, and avoid using overly professional or obscure medical terms. "

[0148] Figure 4 It is a structural schematic diagram of a device for generating a prompt word template according to an embodiment of the present disclosure;

[0149] like Figure 4 As shown, the generating device 400 of the prompt word template includes:

[0150] The acquisition module 401 is used to acquire a first data set and an initial prompt word template corresponding to the target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer;

[0151] A first determination module 402, configured to determine a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description;

[0152] A second determination module 403 is used to determine a first evaluation result corresponding to the first answer according to the reference answer and the first answer;

[0153] The optimization module 404 is used to optimize the initial prompt word template based on the first evaluation result and the text description to obtain a target prompt word template.

[0154] In some embodiments of the present disclosure, a processing module is also included, which is used to:

[0155] Determine a second answer corresponding to the sample data according to the target prompt word template, the sample data and the corresponding text description;

[0156] Determine a second evaluation result corresponding to the second answer according to the second answer and the reference answer;

[0157] Determine the prompt word template to be optimized from the target prompt word template and the initial prompt word template;

[0158] Based on the second evaluation result and the text description, the prompt word template to be optimized is optimized to obtain an optimized target prompt word template.

[0159] In some embodiments of the present disclosure, the processing module is used to:

[0160] Determine a first score corresponding to the initial prompt word template and a second score corresponding to the target prompt word template;

[0161] The prompt word template with the highest score between the first score and the second score is determined as the prompt word template to be optimized.

[0162] In some embodiments of the present disclosure, the processing module is used to:

[0163] Determine a third score corresponding to each first answer according to the difference between the first answer corresponding to each sample data and the reference answer;

[0164] Determine the first score corresponding to the initial prompt word template according to the third score corresponding to each first answer;

[0165] Determine a fourth score corresponding to each second answer according to the difference between the second answer corresponding to each sample data and the reference answer;

[0166] The second score corresponding to the target prompt word template is determined according to the fourth score corresponding to each second answer.

[0167] In some embodiments of the present disclosure, the processing module is used to:

[0168] In the case where the prompt word template to be optimized is the initial prompt word template, the first evaluation result and the second evaluation result corresponding to the sample data are merged to obtain a third evaluation result corresponding to the sample data;

[0169] Based on the third evaluation result and the text description, the initial prompt word is optimized to obtain an optimized target prompt word template.

[0170] In some embodiments of the present disclosure, the first determining module 402 is configured to:

[0171] Fill the text description into the preset position in the initial prompt word template to obtain the first prompt word corresponding to the sample data;

[0172] The first prompt word and sample data are input into the first large model to obtain a first answer corresponding to the sample data.

[0173] In some embodiments of the present disclosure, the second determining module 403 is used to:

[0174] When the target task is a preset task, the reference answer, the first answer and the second prompt word are input into the second largest model to obtain the first evaluation result, wherein the second prompt word is used to prompt the second largest model to evaluate the first answer based on the reference answer.

[0175] In some embodiments of the present disclosure, the optimization module 404 is used to:

[0176] Determining target sample data for which the first evaluation result indicates error;

[0177] In the case where there are at least two target sample data, determining the text description associated with the first evaluation result based on the first evaluation result and the text description corresponding to the target sample data;

[0178] The initial prompt word template, the third prompt word, the first evaluation result, and the text description associated with the first evaluation result are input into the third largest model to obtain the target prompt word template, wherein the third prompt word is used to prompt the third largest model to optimize the initial prompt word template based on the first evaluation result and the corresponding text description.

[0179] In some embodiments of the present disclosure, the optimization module 404 is used to:

[0180] Determining target sample data for which the first evaluation result indicates error;

[0181] When there are at least two target sample data, clustering the first evaluation results corresponding to the target sample data to determine the target evaluation results corresponding to the initial prompt word template;

[0182] Determine a text description associated with the target evaluation result based on the target sample data to which the first evaluation result corresponding to the target evaluation result belongs and the text description corresponding to the target sample data;

[0183] The initial prompt word template, the third prompt word, the target evaluation result, and the text description associated with the target evaluation result are input into the third language model to obtain the target prompt word template, wherein the third prompt word is used to prompt the third language model to optimize the initial prompt word template based on the target evaluation result and the associated text description.

[0184] In some embodiments of the present disclosure, the sample data is at least one of the following:

[0185] Text data, image data and audio data.

[0186] It should be noted that the above explanation of the method for generating a prompt word template is also applicable to the device for generating a prompt word template in this embodiment, and will not be repeated here.

[0187] In the disclosed embodiment, the first answer corresponding to the sample data is determined according to the initial prompt word template, the sample data and the corresponding text description, and then the first evaluation result corresponding to the first answer is determined according to the reference answer corresponding to the sample data and the first answer, and finally the initial prompt word template is optimized based on the first evaluation result and the text description to obtain the target prompt word template. Thus, the defects of the initial prompt word template can be determined by the difference between the first answer corresponding to the sample data under the initial prompt word template and the reference answer, and then the initial prompt word template is optimized in combination with the text description corresponding to the sample data, thereby realizing the automatic optimization of the prompt word template, improving the efficiency and accuracy of the optimization of the prompt word template, and thus improving the quality of the optimized target prompt word template.

[0188] 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.

[0189] Figure 5 A schematic block diagram of an example electronic device 500 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 processing, 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.

[0190] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0191] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0192] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 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, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the generation method of the prompt word template. For example, in some embodiments, the generation method of the prompt word template may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the generation method of the prompt word template described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the generation method of the prompt word template in any other appropriate manner (e.g., by means of firmware).

[0193] Various implementations 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), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations 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.

[0194] 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.

[0195] 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, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, 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 (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or 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.

[0196] 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 CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and 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).

[0197] The systems and techniques described herein can be implemented in a computing system that includes backend 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 frontend 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 backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0198] 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 between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0199] 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.

[0200] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. In the description of the present disclosure, the words "if" and "if" used can be interpreted as "at the time of" or "when" or "in response to determination" or "under the circumstances of".

[0201] 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 spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for generating a prompt word template, wherein: include: Obtaining a first data set and an initial prompt word template corresponding to the target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer; Determine a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description; Determine a first evaluation result corresponding to the first answer according to the reference answer and the first answer; Based on the first evaluation result and the text description, the initial prompt word template is optimized to obtain a target prompt word template.

2. The method according to claim 1, wherein: After optimizing the initial prompt word template based on the first evaluation result and the text description to obtain a target prompt word template, the method further includes: Determining a second answer corresponding to the sample data according to the target prompt word template, the sample data and the corresponding text description; Determine a second evaluation result corresponding to the second answer according to the second answer and the reference answer; Determining a prompt word template to be optimized from the target prompt word template and the initial prompt word template; Based on the second evaluation result and the text description, the prompt word template to be optimized is optimized to obtain an optimized target prompt word template.

3. The method according to claim 2, wherein: The step of determining the prompt word template to be optimized from the target prompt word template and the initial prompt word template includes: Determine a first score corresponding to the initial prompt word template and a second score corresponding to the target prompt word template; The prompt word template with the highest score between the first score and the second score is determined as the prompt word template to be optimized.

4. The method according to claim 3, wherein: The determining of the first score corresponding to the initial prompt word template and the second score corresponding to the target prompt word template includes: Determine a third score corresponding to each of the first answers according to the difference between the first answer corresponding to each of the sample data and the reference answer; Determining a first score corresponding to the initial prompt word template according to a third score corresponding to each of the first answers; Determine a fourth score corresponding to each of the second answers according to a difference between the second answer corresponding to each of the sample data and the reference answer; The second score corresponding to the target prompt word template is determined according to the fourth score corresponding to each of the second answers.

5. The method according to claim 2, wherein: After determining the prompt word template to be optimized from the target prompt word template and the initial prompt word template, the method further includes: In a case where the prompt word template to be optimized is the initial prompt word template, merging the first evaluation result and the second evaluation result corresponding to the sample data to obtain a third evaluation result corresponding to the sample data; Based on the third evaluation result and the text description, the initial prompt word is optimized to obtain an optimized target prompt word template.

6. The method according to claim 1, wherein: Determining a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description includes: Fill the text description into a preset position in the initial prompt word template to obtain a first prompt word corresponding to the sample data; The first prompt word and the sample data are input into a first large model to obtain a first answer corresponding to the sample data.

7. The method according to claim 1, wherein: Determining a first evaluation result corresponding to the first answer according to the reference answer and the first answer includes: When the target task is a preset task, the reference answer, the first answer and the second prompt word are input into the second largest model to obtain the first evaluation result, wherein the second prompt word is used to prompt the second largest model to evaluate the first answer based on the reference answer.

8. The method according to claim 1, wherein: The step of optimizing the initial prompt word template based on the first evaluation result and the text description to obtain a target prompt word template includes: Determining target sample data for which the first evaluation result indicates error; In a case where there are at least two target sample data, determining a text description associated with the first evaluation result based on the first evaluation result and the text description corresponding to the target sample data; The initial prompt word template, the third prompt word, the first evaluation result, and the text description associated with the first evaluation result are input into the third largest model to obtain the target prompt word template, wherein the third prompt word is used to prompt the third largest model to optimize the initial prompt word template based on the first evaluation result and the corresponding text description.

9. The method according to claim 1, characterized in that: The step of optimizing the initial prompt word template based on the first evaluation result and the text description to obtain a target prompt word template includes: Determining target sample data for which the first evaluation result indicates error; In the case where there are at least two target sample data, clustering the first evaluation results corresponding to the target sample data to determine the target evaluation result corresponding to the initial prompt word template; Determining a text description associated with the target evaluation result based on the target sample data to which the first evaluation result corresponding to the target evaluation result belongs and the text description corresponding to the target sample data; The initial prompt word template, the third prompt word, the target evaluation result, and the text description associated with the target evaluation result are input into the third largest language model to obtain the target prompt word template, wherein the third prompt word is used to prompt the third largest language model to optimize the initial prompt word template based on the target evaluation result and the associated text description.

10. The method according to any one of claims 1 to 9, wherein: The sample data is at least one of the following: Text data, image data and audio data.

11. A device for generating a prompt word template, wherein: include: An acquisition module, used to acquire a first data set and an initial prompt word template corresponding to a target task, wherein the first data set includes sample data, a text description corresponding to the sample data, and a reference answer; A first determination module, configured to determine a first answer corresponding to the sample data according to the initial prompt word template, the sample data and the corresponding text description; A second determination module, configured to determine a first evaluation result corresponding to the first answer according to the reference answer and the first answer; An optimization module is used to optimize the initial prompt word template based on the first evaluation result and the text description to obtain a target prompt word template.

12. The device according to claim 11, wherein Also included is a processing module for: Determining a second answer corresponding to the sample data according to the target prompt word template, the sample data and the corresponding text description; Determine a second evaluation result corresponding to the second answer according to the second answer and the reference answer; Determining a prompt word template to be optimized from the target prompt word template and the initial prompt word template; Based on the second evaluation result and the text description, the prompt word template to be optimized is optimized to obtain an optimized target prompt word template.

13. The device according to claim 12, wherein: The processing module is used to: Determine a first score corresponding to the initial prompt word template and a second score corresponding to the target prompt word template; The prompt word template with the highest score between the first score and the second score is determined as the prompt word template to be optimized.

14. The device according to claim 13, wherein: The processing module is used to: Determine a third score corresponding to each of the first answers according to the difference between the first answer corresponding to each of the sample data and the reference answer; Determining a first score corresponding to the initial prompt word template according to a third score corresponding to each of the first answers; Determine a fourth score corresponding to each of the second answers according to a difference between the second answer corresponding to each of the sample data and the reference answer; The second score corresponding to the target prompt word template is determined according to the fourth score corresponding to each of the second answers.

15. The device according to claim 12, wherein: The processing module is used to: In a case where the prompt word template to be optimized is the initial prompt word template, merging the first evaluation result and the second evaluation result corresponding to the sample data to obtain a third evaluation result corresponding to the sample data; Based on the third evaluation result and the text description, the initial prompt word is optimized to obtain an optimized target prompt word template.

16. The device according to claim 11, wherein The first determining module is used to: Fill the text description into a preset position in the initial prompt word template to obtain a first prompt word corresponding to the sample data; The first prompt word and the sample data are input into a first large model to obtain a first answer corresponding to the sample data.

17. The device according to claim 11, wherein: The second determining module is used to: When the target task is a preset task, the reference answer, the first answer and the second prompt word are input into the second largest model to obtain the first evaluation result, wherein the second prompt word is used to prompt the second largest model to evaluate the first answer based on the reference answer.

18. The device according to claim 11, wherein The optimization module is used to: Determining target sample data for which the first evaluation result indicates error; In a case where there are at least two target sample data, determining a text description associated with the first evaluation result based on the first evaluation result and the text description corresponding to the target sample data; The initial prompt word template, the third prompt word, the first evaluation result, and the text description associated with the first evaluation result are input into the third largest model to obtain the target prompt word template, wherein the third prompt word is used to prompt the third largest model to optimize the initial prompt word template based on the first evaluation result and the corresponding text description.

19. The device according to claim 11, characterized in that The optimization module is used to: Determining target sample data for which the first evaluation result indicates error; In the case where there are at least two target sample data, clustering the first evaluation results corresponding to the target sample data to determine the target evaluation result corresponding to the initial prompt word template; Determining a text description associated with the target evaluation result based on the target sample data to which the first evaluation result corresponding to the target evaluation result belongs and the text description corresponding to the target sample data; The initial prompt word template, the third prompt word, the target evaluation result, and the text description associated with the target evaluation result are input into the third largest language model to obtain the target prompt word template, wherein the third prompt word is used to prompt the third largest language model to optimize the initial prompt word template based on the target evaluation result and the associated text description.

20. The device according to any one of claims 11 to 19, wherein: The sample data is at least one of the following: Text data, image data and audio data.

21. 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 10.

22. 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 to 10.

23. A computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 10.

Images