A prompt word optimization method, device, equipment, medium and product

Abstract

The present application provides a method, apparatus, device, medium, and product for optimizing prompt words. The method includes: obtaining at least one first prompt word associated with a target business, and obtaining a business data sample set related to the target business; for each of the at least one first prompt word, performing the following steps: obtaining a first evaluation result of the first prompt word on the business data sample set, and based on the first evaluation result, determining error sample information corresponding to the first prompt word; sending a target prompt word for optimizing the first prompt word to a first language model; and obtaining a second prompt word output by the first language model. In this method, the prompt word to be optimized is evaluated for its task processing effect in the business data sample set, and combined with relevant information about the error data sample, the prompt word is optimized using a language model to improve the prompting ability of the optimized prompt word in the target business, thereby optimizing the prompting effect of the prompt word.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a prompt word optimization method, device, electronic device, computer-readable storage medium, and computer program product. Background Art

[0002] With the continuous development of computer technology and natural language processing (NLP) technology, the scale of language models has gradually expanded. In order to obtain better model reasoning effects, prompt learning came into being.

[0003] Prompt learning refers to modifying downstream tasks through prompt words without changing the parameters of the pre-trained language model, so that the pre-trained language model can better understand the downstream tasks and thus achieve better model inference results.

[0004] In prompt learning, the design of prompt words is particularly important. In related technologies, technicians can use language models to automatically optimize prompt words. However, this method is difficult to meet the task processing requirements of different businesses. Summary of the Invention

[0005] This application provides a method for optimizing prompt words. This method can improve the prompt effect of prompt words in specific services. This application also provides a device, electronic device, computer-readable storage medium, and computer program product corresponding to the above method.

[0006] In a first aspect, the present application provides a method for optimizing prompt words, the method comprising:

[0007] Acquire at least one first prompt word associated with a target business, and acquire a business data sample set related to the target business; wherein the data samples in the business data sample set include annotation results;

[0008] For each first prompt word in the at least one first prompt word, perform the following steps:

[0009] Obtain a first evaluation result of the first prompt word on the business data sample set, and determine error sample information corresponding to the first prompt word based on the first evaluation result; send a target prompt word for optimizing the first prompt word to a first language model, wherein the target prompt word includes the first prompt word and the error sample information corresponding to the first prompt word; and obtain a second prompt word output by the first language model.

[0010] In a second aspect, the present application provides a device for optimizing prompt words, the device comprising:

[0011] an acquisition module, configured to acquire at least one first prompt word associated with a target business, and acquire a business data sample set related to the target business; wherein the data samples in the business data sample set include annotation results;

[0012] An execution module is configured to perform the following steps for each first prompt word of the at least one first prompt word: obtaining a first evaluation result of the first prompt word on the business data sample set, and determining error sample information corresponding to the first prompt word based on the first evaluation result; sending a target prompt word for optimizing the first prompt word to a first language model, the target prompt word including the first prompt word and the error sample information corresponding to the first prompt word; and obtaining a second prompt word output by the first language model.

[0013] In a third aspect, the present application provides an electronic device comprising a processor and a memory. The processor and the memory communicate with each other. The processor is configured to execute instructions stored in the memory to cause the electronic device to perform the method for optimizing prompt words as described in the first aspect or any implementation of the first aspect.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions instruct an electronic device to execute the method for optimizing prompt words described in the first aspect or any one of the implementations of the first aspect.

[0015] In a fifth aspect, the present application provides a computer program product comprising instructions, which, when executed on an electronic device, enables the electronic device to execute the method for optimizing prompt words as described in the first aspect or any one of the implementations of the first aspect.

[0016] Based on the implementation methods provided in the above aspects, this application can also be further combined to provide more implementation methods.

[0017] It can be seen from the above technical solutions that this application has the following advantages:

[0018] The present application provides a method for optimizing prompt words. The method first obtains at least one first prompt word associated with a target business, and obtains a business data sample set related to the target business, wherein the data samples in the business data sample set include annotation results. Then, for each first prompt word in the at least one first prompt word, the following steps are performed: obtaining a first evaluation result of the first prompt word on the business data sample set; based on the first evaluation result, determining error sample information corresponding to the first prompt word; sending a target prompt word for optimizing the first prompt word to a first language model, wherein the target prompt word includes the first prompt word and the error sample information corresponding to the first prompt word; and obtaining a second prompt word output by the first language model.

[0019] In this method, the task processing performance of the prompt word to be optimized (i.e., the first prompt word) is evaluated on a set of business data samples. By comparing the annotation results with the evaluation results, erroneous data samples with incorrect judgments are identified. Combined with the relevant information of the erroneous data samples, the prompt word is optimized using a language model. This improves the prompting ability of the optimized prompt word (i.e., the second prompt word) in the target business and optimizes its prompting effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical methods of the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments.

[0021] Figure 1 A flowchart of a method for optimizing prompt words provided in an embodiment of the present application;

[0022] Figure 2 A flowchart of a method for optimizing prompt words provided in an embodiment of the present application;

[0023] Figure 3 A flowchart of a method for optimizing prompt words provided in an embodiment of the present application;

[0024] Figure 4 A schematic diagram of the structure of a prompt word optimization device provided in an embodiment of the present application;

[0025] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The terms "first" and "second" in the embodiments of this application are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Therefore, features specified as "first" or "second" may explicitly or implicitly include one or more of the features.

[0027] First, some technical terms and application scenarios involved in the embodiments of this application are introduced.

[0028] Natural language processing (NLP) is a machine learning technique used to enable computing devices to interpret, process, and understand human language. Based on NLP, computing devices can perform a variety of tasks, including but not limited to part-of-speech tagging, word sense disambiguation, speech recognition, machine translation, and sentiment analysis.

[0029] With the continuous development of natural language processing technology, the industry has seen the emergence of a "pre-training + prompt + prediction" natural language processing paradigm. This "pre-training + prompt + prediction" paradigm uses a pre-trained language model based on prompt learning for natural language processing. This eliminates the need to fine-tune the parameters of the pre-trained language model before applying it to downstream tasks. Instead of modifying the parameters of the pre-trained language model, the downstream task is modified using prompts, enabling the pre-trained language model to better understand the downstream task.

[0030] Specifically, the prompt word usually includes a placeholder, and the user's input text or other text is filled into the placeholder of the prompt word to form a complete prompt information.

[0031] Taking the downstream task of "text classification" as an example, the pre-trained language model needs to analyze the user's input text and output the classification result of the input text. For example, the user's input text is "The plot is smooth, the actors' acting is good, and the special effects are also good," and the prompt word is "The user's evaluation of this movie is {{xxx}}." The prompt word prompts the pre-trained language model to perform text classification, allowing it to output the classification result of the input text (such as positive or negative review category). This allows the pre-trained language model to better understand the downstream task and generate accurate model output.

[0032] Considering the increasing amount of training data and continuous breakthroughs in model ontology technology, the capabilities of pre-trained language models are also continuously improving, and the prompt words should also be updated and optimized accordingly. In related technologies, the methods for optimizing prompt words are mainly divided into manual optimization and automatic optimization.

[0033] Manual optimization of prompt words can be understood as manual adjustment of prompt words by technical personnel. Manual optimization of prompt words is inefficient, and the optimization experience of technical personnel is not universal enough. When the language model is upgraded or switched, the original optimization experience may not be applicable.

[0034] In the automatic optimization of prompt words, a language model is usually used as a prompt word generator. For example, the prompt word is scored using the language model, and the score and the prompt word are then input into the language model. The language model then combines the score and generates an optimized prompt word.

[0035] However, the above-mentioned automatic optimization method of prompt words is difficult to meet the task processing requirements of different businesses, and the optimization of prompt words lacks pertinence.

[0036] In view of this, the present application provides a method for optimizing prompt words. The method first obtains at least one first prompt word associated with a target business, and obtains a business data sample set related to the target business, wherein the data samples in the business data sample set include annotation results. Then, for each first prompt word in the at least one first prompt word, the following steps are performed: obtaining a first evaluation result of the first prompt word on the business data sample set; based on the first evaluation result, determining error sample information corresponding to the first prompt word; sending a target prompt word for optimizing the first prompt word to a first language model, wherein the target prompt word includes the first prompt word and the error sample information corresponding to the first prompt word; and obtaining a second prompt word output by the first language model.

[0037] In this method, the task processing performance of the prompt word to be optimized (i.e., the first prompt word) is evaluated on a set of business data samples. By comparing the annotation results with the evaluation results, erroneous data samples with incorrect judgments are identified. Combined with the relevant information of the erroneous data samples, the prompt word is optimized using a language model. This improves the prompting ability of the optimized prompt word (i.e., the second prompt word) in the target business and optimizes its prompting effectiveness.

[0038] To facilitate understanding of the technical solutions provided in the embodiments of the present application, they will be described below with reference to the accompanying drawings.

[0039] See also Figure 1 The flowchart of a method for optimizing prompt words provided in an embodiment of the present application is shown, and the method specifically includes:

[0040] S101: Acquire at least one first prompt word associated with a target business, and acquire a business data sample set related to the target business.

[0041] In the embodiment of the present application, the first prompt word can be understood as the prompt word to be optimized, and the target business can be any business associated with the first prompt word.

[0042] The target business may be associated with a target task, wherein the target task may be any natural language processing task. In other words, in the process of processing the target task with the help of a language model, the first prompt word may assist the language model in processing the target task.

[0043] Different target businesses can correspond to different target tasks, and thus different first prompts. For example, if the target business is customer service, the target task could be to judge the customer service quality based on the conversation content. The first prompt could be, "The conversation content between the customer service representative and the customer is {{xxx}}. Please rate the customer service quality." For another example, if the target business is document management, the target task could be to score documents. The first prompt could be, "The document content is {{xxx}}. Please rate the document."

[0044] A business data sample set typically includes multiple data samples, and the data samples are related to the target business. In some embodiments, the data samples in the business data sample set can be real business data, such as business data generated during the operation of the target business. In other embodiments, the data samples in the business data sample set can also be generated business data, such as test business data generated in combination with the business logic of the target business, or business data collected through other data collection methods, which is not limited in this embodiment of the present application.

[0045] In an embodiment of the present application, data samples in a business data sample set may be pre-labeled, i.e., the data samples in the business data sample set include labeling results. Typically, the labeling results can be understood as labels, and the labeling results can represent a reference evaluation or benchmark evaluation of the data sample in the target task.

[0046] For example, if the target business is customer service and the target task is to judge customer service quality, the labeling results of the data sample can be "excellent," "good," "qualified," and "unqualified." For another example, if the target business is document management and the target task is to score documents, the labeling results of the data sample can be scores such as "90," "80," "75," and "60."

[0047] The embodiments of the present application do not limit the method of labeling data samples. For example, data samples can be labeled manually (such as business personnel, labeling personnel), or data samples can be automatically labeled by labeling models, labeling tools, etc.

[0048] For each first prompt word in the at least one first prompt word, perform the following S102 to S104:

[0049] S102: Obtain a first evaluation result of the first prompt word on the business data sample set, and determine error sample information corresponding to the first prompt word based on the first evaluation result.

[0050] The first prompt word may be understood as the first prompt word currently being optimized, and the first evaluation result may be understood as the evaluation of the data sample in the target task after the target task is processed with the help of the first prompt word.

[0051] That is to say, in the embodiment of the present application, the first prompt word is actually applied in the target task, and the evaluation result of the data sample after the target task is processed by applying the first prompt word is obtained, and then the prompt ability of the first prompt word is evaluated.

[0052] The above process can be implemented with the help of a second language model. A second language model can be understood as a model that has the ability to understand and generate natural language and can process different natural language tasks. For example, the second language model can be a deep learning model trained using text data.

[0053] In a specific implementation, a first evaluation prompt word for evaluating the first prompt word is sent to the second language model, and a first evaluation result output by the second language model is received, wherein the first evaluation prompt word includes the first prompt word and a business data sample set.

[0054] In other words, by evaluating the prompting capability of the first prompt word, the second language model is instructed to use the first prompt word to process the data samples in the business data sample set for the target task, thereby generating a first evaluation result for the data samples in the business data sample set. In this way, the prompting capability of the first prompt word in actual business operations is evaluated.

[0055] The error sample information can be understood as information related to the error data sample. Specifically, the error sample information can include the error data sample, the labeling result of the error data sample, and the first evaluation result of the error data sample.

[0056] Incorrect data samples may be data samples where the first evaluation result is inconsistent with the labeling result. In other words, incorrect data samples refer to data samples where, when the target task is processed using the first prompt word, the language model outputs an evaluation that differs from the reference evaluation. In other words, incorrect data samples result in incorrect judgments based on the prompt capability of the first prompt word.

[0057] In the embodiment of the present application, error data samples, labeling results of the error data samples, and first evaluation results of the error data samples are collected so as to subsequently optimize the first prompt word in combination with the error sample information.

[0058] In some possible implementations, error sample information may also be output by the second language model. Specifically, the first evaluation prompt may also indicate how the error data sample was determined and the content of the error sample information. Thus, by leveraging the prompting capabilities of the first evaluation prompt, the second language model may also output the error sample information after generating the first evaluation result.

[0059] It should be noted that the above process can also be implemented in other ways, for example, by using a machine learning algorithm to obtain the first evaluation result and error sample information.

[0060] In some embodiments, the data samples in the business data sample set may further include a labeling reason. The labeling reason may indicate the reason for generating the labeling result. In other words, the labeling reason may be used to explain the labeling result of the data sample in the business data sample set.

[0061] The reason for labeling can be presented in the form of business rules. For example, if the target business is customer service, and a data sample is labeled "unqualified," the reason could be "Customer service did not follow up." Another example is if the target business is document management, and a data sample is labeled "60," the reason could be "Document content format error."

[0062] When the error sample information includes multiple error data samples, the error data samples may be sorted. Specifically, the error data samples may be sorted based on the degree of difference between the first evaluation results and the annotation results of the error data samples, and / or the number of occurrences of the annotation reasons of the error data samples in the business data sample set, to obtain a sorting result for the error data samples. In the error sample information corresponding to the first prompt word, the error sample data may be arranged according to the sorting result for the error data samples.

[0063] The difference between the first evaluation result and the labeling result can be understood as the prediction error of the target task using the first prompt word. For example, when the task type indicated by the first prompt word (i.e., the task type of the target task) is a regression task, the labeling result of the incorrect data sample is 90, the first evaluation result of the incorrect data sample is 70, and the difference between the first evaluation result and the labeling result is 20.

[0064] The number of times a labeling reason for an erroneous data sample occurs in the business data sample set can be understood as the number of times that reason occurs in each data sample in the business data sample set. For example, if there are 100 data samples in the business data sample set, and the labeling reason for an erroneous data sample is labeling reason A, 40 of the 100 data samples have labeling reason A. In this case, the number of times labeling reason A occurs in the business data sample set is 40.

[0065] In the embodiment of the present application, the error sample information is integrated based on the sorting results of multiple error data samples to form an error data sample sequence in the error sample information. In the error data sample sequence, information related to the error data sample is arranged according to the sorting results.

[0066] For example, the error data samples include error data sample A, error data sample B and error data sample C, and the sorting results are error data sample B, error data sample A, error data sample C. At this time, the error data sample sequence in the error sample information can be: {error data sample B, the labeling result of error data sample B, the first sorting result of error data sample B}, {error data sample A, the labeling result of error data sample A, the first sorting result of error data sample A}, {error data sample C, the labeling result of error data sample C, the first sorting result of error data sample C}.

[0067] In some possible implementations, the greater the difference between the first evaluation result and the labeling result, the higher the ranking of the erroneous data sample. The more frequently the labeling reason for the erroneous data sample appears in the business data sample set, the higher the ranking of the erroneous data sample. Thus, when subsequently optimizing the first prompt word, focus is placed on erroneous sample information related to erroneous data samples with large differences and / or erroneous sample information related to erroneous data samples with a high number of labeling reasons, further improving the optimization effect.

[0068] S103: Sending a target prompt word for optimizing the first prompt word to the first language model.

[0069] S104: Obtain a second prompt word output by the first language model.

[0070] In the embodiments of the present application, the prompt word is optimized using a first language model. The first language model can be understood as a language model used to optimize the prompt word. The first language model can be understood as a model that has natural language understanding and generation capabilities and can process different natural language tasks. For example, the first language model can be a deep learning model trained using text data.

[0071] It should be noted that the first language model and the second language model can be the same language model or different language models. Furthermore, the first language model and the second language model can be flexibly selected based on business needs. The first language model and the second language model can be publicly deployed or privately deployed, which is not limited in this embodiment of the present application.

[0072] The target prompt word can be understood as the prompt word used to optimize the first prompt word. In other words, the prompting ability of the target prompt word instructs the first language model to optimize the first prompt word based on error sample information, generating an optimized second prompt word. In this way, the first language model learns and understands error sample information, and optimizes the prompt word based on a sample set of business data, making the optimized second prompt word more relevant to business practices and providing better prompting ability.

[0073] In some possible implementations, the target prompt word may be presented in the following form:

[0074] You are a professional prompt optimization expert. You are responsible for reflecting on incorrect sample judgment results based on given prompts and language models. You need to carefully analyze the given prompts and why they produce incorrect results for given business data samples. Through analysis and reflection, you need to output optimized prompts with structured information such as roles, goals, and key business rules. This optimized prompt can better achieve correct judgment results for given samples.

[0075] The input includes prompt words and error sample information. The error sample information contains a series of information related to the error data sample. Each information related to the error data sample consists of the error data sample, the evaluation result of the error data sample, the annotation result, and the annotation reason of the annotation result. Different samples represent the conversation content of different engineers.

[0076] Prompt words:

[0077] {{prompt}}

[0078] Sample error information:

[0079] {{examples}}

[0080] Please combine the above information and directly output the optimized prompt words with structured information."

[0081] In this method, the task processing performance of the prompt word to be optimized (i.e., the first prompt word) is evaluated on a set of business data samples. By comparing the annotation results with the evaluation results, erroneous data samples with incorrect judgments are identified. Combined with the relevant information of the erroneous data samples, the prompt word is optimized using a language model. This improves the prompting ability of the optimized prompt word (i.e., the second prompt word) in the target business and optimizes its prompting effectiveness.

[0082] The above article describes the optimization process for the first prompt word. After the optimization of the first prompt word is completed, the optimization effect can be further evaluated.

[0083] In some embodiments, the second prompt word includes multiple prompt words. For example, when the first prompt word includes multiple prompt words, batch optimization is performed on the multiple first prompt words to obtain multiple second prompt words.

[0084] Since the optimized second prompt words have not been processed for the target task, their prompting capabilities are unknown. Therefore, the second prompt words can be used in the target task. Specifically, the optimization effect of each second prompt word is determined, and based on the optimization effect of each second prompt word, a third prompt word is determined from at least one second prompt word.

[0085] In the embodiment of the present application, the optimization effect can be determined based on the second evaluation result and the labeling result of the business data sample set. The second evaluation result of the business data sample set can be understood as the evaluation of the data sample in the target task after the target task is processed with the help of the second prompt word.

[0086] That is, the optimized second prompt word is actually applied in the target task, and the prompt ability of the second prompt word is evaluated through the evaluation results of the business data sample after the second prompt word is applied to process the target task, so as to determine the optimization effect of the prompt word.

[0087] Specifically, the optimization effect can be the ratio of the number of data samples whose second evaluation results and annotation results meet the set conditions to the number of business data samples in the business data sample set. That is, the optimization effect can be expressed in the form of a score. For example, if the number of data samples whose second evaluation results and annotation results meet the set conditions is 80, and the number of data samples in the business data sample set is 100, the optimization effect can be 0.8.

[0088] Different target tasks can have different set conditions. For example, when the target task type is a classification task, the set condition can be that the second evaluation result is the same as the labeling result. For another example, when the target task type is a regression task, the set condition can be that the difference between the second evaluation result and the labeling result is less than a difference threshold.

[0089] Similarly, the above process can also be implemented using a second language model. Specifically, for each second prompt word, the following steps are performed: sending a second evaluation prompt word to the second language model for evaluating the optimization effect of the second prompt word, and receiving the optimization effect output by the second language model. The second evaluation prompt word includes the second prompt word, a sample set of business data, and a method for determining the optimization effect.

[0090] In other words, based on the second evaluation prompt word's prompting ability, the second language model is instructed to use the second prompt word to process the target task on the data samples in the business data sample set, generate a second evaluation result for the data samples in the business data sample set, and determine the optimization effect based on the second evaluation result. In this way, the prompting ability of the second prompt word in actual business operations is evaluated, and the optimization effect of the prompt word is determined.

[0091] After determining the optimization effect of the second prompt word, the second prompt words can be screened to determine the third prompt word. For example, when the optimization effect is expressed in the form of a score, the optimization effect of the second prompt words can be ranked from high to low, and the top-K second prompt words can be selected as the third prompt word.

[0092] In this way, after the prompt word optimization is completed, the prompt ability of the optimized prompt words is evaluated in actual business, and the prompt words that perform well in actual business are selected for actual use, thereby ensuring the effectiveness of the optimized prompt words in actual business.

[0093] Based on the optimization method of prompt words described above, the embodiment of the present application can also provide a framework for optimizing prompt words. Figure 2 As shown, the first prompt word can be generated from the initial prompt word. The second language model is used to evaluate the prompting ability of the first prompt word to obtain error sample information. Then, based on the error sample information, the first prompt word is optimized using the first language model to generate the second prompt word. The second language model is then used to evaluate the prompting ability of the second prompt word to obtain the optimization result. Then, based on the optimization result, the third prompt word is selected from the second prompt word.

[0094] In some embodiments, there are multiple rounds of optimization for the prompt word. Specifically, the third prompt word can be determined as the first prompt word in the next round of optimization, so that the third prompt word can be optimized in the next round of optimization.

[0095] That is, after optimizing the first prompt word and selecting a third prompt word with better prompting ability from the optimized second prompt word, the third prompt word can be further optimized to enhance the relevance of the prompt word to the target business and continue to improve the prompting ability of the prompt word.

[0096] It should be noted that the embodiments of the present application do not limit the number of iterative cycles of the optimization process. For example, a loop termination condition can be pre-set, and after the loop termination condition is reached, the prompt word generated in the last round of optimization process is used as the final prompt word.

[0097] The following describes the process of generating the first prompt word from the initial prompt word.

[0098] In actual applications, the number of initial prompt words associated with the target business may be small, for example, there may be only one initial prompt word associated with the target business. In this case, the first prompt word can be generated from the initial prompt word in various ways. Based on the preliminary optimization of the prompt words, the number of prompt words can be increased to achieve better optimization results in the future.

[0099] In some possible implementations, the process of generating the first prompt word from the initial prompt word is implemented using a third language model. The third language model can be understood as a language model used to initially optimize the prompt word. The third language model can be understood as a model that has natural language understanding and generation capabilities and can process different natural language tasks. For example, the third language model can be a deep learning model trained using text data.

[0100] Similarly, the third language model may be the same language model as the first language model and the second language model, or may be a different language model from the first language model or the second language model, and this embodiment of the present application does not impose any limitation on this.

[0101] In a specific implementation, an initial prompt word is first obtained, an optimized prompt word for optimizing the initial prompt word is sent to a third language model, and at least one first prompt word output by the third language model is received, wherein the optimized prompt word includes the initial prompt word.

[0102] By optimizing the prompting power of the prompt word, the third language model is instructed to optimize the initial prompt word to generate the first prompt word. In this way, the initial prompt word of low quality is preliminarily optimized, making the generated first prompt word logically coherent and enhancing the prompting power of the prompt word to a certain extent, improving the optimization efficiency of subsequent prompt words.

[0103] In some embodiments, the optimization prompt words may be presented in the following forms:

[0104] "You are a professional prompt word optimization master. Please fully understand the input prompt word and then output the optimized prompt word with structured information such as roles, goals, and key business rules.

[0105] The prompt words to be entered are as follows:

[0106] {{prompt}}

[0107] Please output optimized prompt words with structured information."

[0108] In other embodiments, the data samples in the business data sample set include annotation reasons. In this case, the optimization prompt word can also include the annotation reason. In this way, when the third language model optimizes the initial prompt word, it can learn the business rules related to the target task by optimizing the annotation reasons in the prompt word. The optimization prompt word can be presented in the following form:

[0109] "You are a professional prompt word optimization expert. Please thoroughly analyze the given prompt word and common annotation reasons. The annotation reasons will be provided along with the number of annotations. The more annotations there are, the more important the corresponding annotation reason is, and the more important it needs to be considered when generating optimized prompt words. Please thoroughly and comprehensively analyze the given prompt word, the annotation reason, and the corresponding number of annotations, and then output optimized prompt words with structured information such as role, goal, and key business rules.

[0110] The prompt words to be entered are as follows:

[0111] {{prompt}}

[0112] Common reasons for annotation are as follows:

[0113] {{reasons}}

[0114] By combining the above information, optimized prompt words with structured information such as roles, goals, and key business rules are directly output.

[0115] In other possible implementations, the first prompt word can be generated without the aid of a language model. Specifically, an initial prompt word is first obtained, and then at least one target data sample is determined from a set of business data samples. The initial prompt word and the at least one target data sample are then concatenated to obtain at least one first prompt word.

[0116] That is, some target data samples are selected from the business data sample set, and based on the few-shot technique, the first prompt word is directly formed using the target data samples and the initial prompt word. For example, the target data sample can be added to the initial prompt word to form the first prompt word.

[0117] Since the target data sample is associated with the target business, and the initial business data sample includes the annotation result, the first prompt word generated using the target data sample and the initial prompt word has the ability to process the target task to a certain extent.

[0118] The embodiments of the present application support screening target data samples in different ways. In some possible implementations, based on the semantic information of the data samples in the business data sample set, the data samples in the business data sample set are clustered, a clustering result is determined, and at least one target data sample is determined based on the clustering result.

[0119] In other words, data samples are screened based on the semantics of the data samples in the business data sample set. For example, semantic information can be represented by a semantic vector. The m data samples closest to the cluster center are selected from the clustering results. Then, data samples with different annotations from these m data samples and closest to the cluster center are selected, resulting in a total of 2m data samples. Duplicates are then removed from these 2m data samples to select the target business data sample. Here, m is a natural number greater than 0.

[0120] In some other possible implementations, the data samples in the business data sample set further include marking reasons, and at least one target data sample is determined based on the number of occurrences of the marking reasons in the business data sample set.

[0121] In other words, data samples are screened based on their labeling reasons. For example, the number of occurrences of each labeling reason in the business data sample set is counted, and the data samples with the top n labeling reasons are selected as target business data samples.

[0122] The framework for optimizing prompt words provided in embodiments of the present application can be applied to both single prompt word optimization and batch optimization of multiple prompt words. In some embodiments, the initial prompt word may include only one prompt word. Preliminary optimization is performed on the initial prompt word to obtain a first prompt word. The first prompt word is evaluated against a set of business data samples using a second language model to obtain a first evaluation result. Based on the first evaluation result, error sample information is determined. Subsequently, the first prompt word is optimized using the first language model in combination with the error sample information to generate a second prompt word.

[0123] In other embodiments, the initial prompt word may also include multiple prompt words. Figure 3 As shown, a preliminary optimization is performed on each of the multiple initial prompt words to obtain multiple first prompt words that are initially optimized from each initial prompt word. Next, a second language model is used to evaluate each first prompt word within a set of business data samples, obtaining a first evaluation result and corresponding error sample information for each first prompt word. Using the first language model and the error sample information corresponding to each first prompt word, each first prompt word is optimized separately to generate a second prompt word that is optimized from each first prompt word. This achieves batch optimization of prompt words and improves prompt word optimization efficiency.

[0124] Furthermore, in the above-mentioned prompt word optimization process, in addition to presenting the optimized prompt word to the user (eg, a business person), intermediate results related to the prompt word optimization may also be presented to the user.

[0125] In some embodiments, the intermediate results include difficult data samples. Specifically, based on the degree of difference between the first evaluation result of the erroneous data sample and the labeled result, the difficult data sample is determined from the erroneous data sample and presented.

[0126] Among them, difficult data samples can be understood as erroneous data samples with a large degree of difference between the first evaluation result and the annotation result, that is, after processing the target task with the help of the first prompt word, it is still difficult to obtain data samples for accurate judgment.

[0127] On the one hand, difficult data samples help optimize prompt words. On the other hand, difficult data samples may also be data samples with labeling errors, such as data samples with incorrect labeling results or incorrect labeling reasons. Therefore, by presenting difficult data samples to users, it helps to optimize business rules and labeling processes.

[0128] In other embodiments, the intermediate result includes a prediction result. Specifically, the first prompt word includes multiple prompt words, and the prediction result is determined based on the first evaluation results of at least some of the first prompt words for the same data sample, and the prediction result is presented.

[0129] Among them, when the task type indicated by the first prompt word (i.e., the task type of the target task) is a classification task, the prediction result is the evaluation result with the largest number of occurrences among the first evaluation results of at least some of the first prompt words for the same data sample. In this case, the number of at least some of the first prompt words needs to be an odd number (2k+1, where k is a natural number greater than 0). For example, the number of first prompt words is 5, and the number of at least some of the first prompt words is 3, including first prompt word A, first prompt word B, and first prompt word C. The first evaluation result of first prompt word A for data sample A is A1, the first evaluation result of first prompt word B for data sample A is A2, and the first evaluation result of first prompt word C for data sample A is A1. In this case, the prediction result is A1.

[0130] When the task type indicated by the first prompt word (i.e., the task type of the target task) is a regression task, the prediction result is the average of the first evaluation results of at least some of the first prompt words for the same data sample. In this case, there is no limit on the number of at least some of the first prompt words. For example, if the number of first prompt words is 5, and the number of at least some of the first prompt words is 2, including first prompt word D and first prompt word E, the first evaluation result of first prompt word D for business data sample B is 70, and the first evaluation result of first prompt word E for business data sample B is 74, then the prediction result is 72.

[0131] By presenting the prediction results to the user, the user can understand the overall judgment of multiple first prompt words, so that the user can filter out prompt words with poor prompt ability and poor prompt effect from multiple first prompt words, thereby improving the optimization efficiency in subsequent rounds of optimization.

[0132] In the embodiment of the present application, high-quality delivery is achieved by presenting difficult data samples and / or prediction results to users, which helps business personnel combine business experience and the optimization process of prompt words to obtain prompt words with better prompt effects.

[0133] Combined with the above Figures 1 to 3 The optimization method of the prompt words provided in the embodiment of the present application is introduced in detail. The following will introduce the device and equipment provided in the embodiment of the present application with reference to the accompanying drawings.

[0134] See also Figure 4 The schematic diagram of the structure of the prompt word optimization device shown in FIG. 40 includes:

[0135] An acquisition module 401 is configured to acquire at least one first prompt word associated with a target business, and acquire a business data sample set related to the target business; wherein the data samples in the business data sample set include annotation results;

[0136] Execution module 402 is configured to perform the following steps for each first prompt word of the at least one first prompt word: obtaining a first evaluation result of the first prompt word on the business data sample set, and determining error sample information corresponding to the first prompt word based on the first evaluation result; sending a target prompt word for optimizing the first prompt word to the first language model, the target prompt word including the first prompt word and the error sample information corresponding to the first prompt word; and obtaining a second prompt word output by the first language model.

[0137] In some possible implementations, the error sample information includes: an error data sample, a labeling result of the error data sample, and a first evaluation result of the error data sample.

[0138] In some possible implementations, the execution module 402 is specifically configured to:

[0139] Sending a first evaluation prompt word for evaluating the first prompt word to a second language model, where the first evaluation prompt word includes the first prompt word and the business data sample set;

[0140] A first evaluation result output by the second language model is received.

[0141] In some possible implementations, the data sample in the business data sample set further includes a marking reason, where the marking reason indicates a reason for generating a marking result; the error sample information includes a plurality of error data samples; and the execution module 402 is further configured to:

[0142] Sorting the plurality of erroneous data samples according to the degree of difference between the first evaluation results and the annotation results of the plurality of erroneous data samples, and / or the number of occurrences of the annotation reasons of the plurality of erroneous data samples in the business data sample set, to obtain a sorting result of the plurality of erroneous data samples;

[0143] In the error sample information corresponding to the first prompt word, the plurality of error sample data are arranged according to the sorting result of the plurality of error data samples.

[0144] In some possible implementations, the obtaining module 401 is specifically configured to:

[0145] Get the initial prompt word;

[0146] Sending an optimized prompt word for optimizing the initial prompt word to a third language model, wherein the optimized prompt word includes the initial prompt word;

[0147] Receive at least one first prompt word output by the third language model.

[0148] In some possible implementations, the obtaining module 401 is specifically configured to:

[0149] Obtaining an initial prompt word, and determining at least one target data sample from the business data sample set;

[0150] The initial prompt word and the at least one target data sample are concatenated to obtain at least one first prompt word.

[0151] In some possible implementations, the data samples in the business data sample set further include a marking reason, where the marking reason indicates a reason for generating a marking result; and the acquisition module 401 is specifically configured to:

[0152] Determine at least one target data sample according to the number of occurrences of the marking reason in the business data sample set; or,

[0153] Based on the semantic information of the data samples in the business data sample set, the data samples in the business data sample set are clustered to determine a clustering result; and according to the clustering result, at least one target data sample is determined.

[0154] In some possible implementations, the apparatus 40 further includes a determining module, configured to:

[0155] Determining an optimization effect of each of the second prompt words;

[0156] A third prompt word is determined from at least one second prompt word according to the optimization effect of each second prompt word.

[0157] In some possible implementations, the determining module is specifically configured to:

[0158] For each second prompt word, perform the following steps:

[0159] Sending a second evaluation prompt word for evaluating an optimization effect of the second prompt word to the second language model, where the second evaluation prompt word includes the second prompt word, a business data sample set, and a method for determining the optimization effect;

[0160] Receive the optimization effect output by the second language model.

[0161] In some possible implementations, the determining module is further configured to:

[0162] The third prompt word is determined as the first prompt word in the next round of prompt word optimization process, so as to optimize the third prompt word in the next round of prompt word optimization process.

[0163] In some possible implementations, the error sample information includes an error data sample, and the apparatus 40 further includes a presentation module, wherein the presentation module is configured to:

[0164] Determining a difficult data sample from the erroneous data sample according to a degree of difference between the first evaluation result and the labeling result of the erroneous data sample;

[0165] The difficult data sample is presented.

[0166] In some possible implementations, the first prompt word includes multiple prompt words, and the presentation module is further configured to:

[0167] determining a prediction result based on first evaluation results of at least part of the first prompt words for the same data sample;

[0168] The prediction results are presented.

[0169] The prompt word optimization device 40 according to the embodiment of the present application may correspond to the method described in the embodiment of the present application, and the above and other operations and / or functions of each module / unit of the prompt word optimization device 40 are respectively to achieve Figures 1 to 3 For the sake of brevity, the corresponding processes of the various methods in the illustrated embodiments are not described again here.

[0170] The embodiment of the present application also provides an electronic device. The electronic device is specifically used to implement Figure 3 The function of the prompt word optimization device 40 in the illustrated embodiment.

[0171] Figure 5 A structural diagram of an electronic device 500 is provided. Figure 5 As shown, the electronic device 500 includes a bus 501, a processor 502, a communication interface 503, and a memory 504. The processor 502, the memory 504, and the communication interface 503 communicate with each other via the bus 501.

[0172] The bus 501 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0173] The processor 502 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0174] The communication interface 503 is used for communicating with the outside, for example, the communication interface 503 can be used for communicating with a terminal.

[0175] The memory 504 may include volatile memory, such as random access memory (RAM), or non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0176] The memory 504 stores executable codes, and the processor 502 executes the executable codes to perform the aforementioned prompt word optimization method.

[0177] Specifically, in the implementation Figure 4 In the case of the embodiment shown, and Figure 4 In the embodiment, each module or unit of the prompt word optimization device 40 is implemented by software. Figure 4 The software or program code required for the functions of each module / unit in the system may be partially or completely stored in the memory 504. The processor 502 executes the program code corresponding to each unit stored in the memory 504 to perform the aforementioned optimization method for the prompt word.

[0178] The present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium capable of being stored by a computing device, or a data storage device such as a data center that contains one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, or a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the aforementioned method for optimizing prompt words applied to the prompt word optimization device 40.

[0179] The present application also provides a computer program product comprising one or more computer instructions that, when loaded and executed on a computing device, fully or partially generate the process or function described in the present application.

[0180] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0181] When the computer program product is executed by a computer, the computer performs any of the aforementioned methods for optimizing prompt words. The computer program product may be a software installation package, which can be downloaded and executed on a computer when any of the aforementioned methods for optimizing prompt words is needed.

[0182] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to the various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the prescribed logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the prescribed function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0183] The units involved in the embodiments described in this application may be implemented in software or hardware, wherein the name of a unit / module does not, in some cases, constitute a limitation on the unit itself.

[0184] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0185] In the context of the present application embodiment, machine-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable medium can 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 can 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.

[0186] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems or devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0187] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0188] It should also be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0189] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0190] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A prompt word optimization method, characterized in that: The method comprises: Acquire at least one first prompt word associated with a target business, and acquire a business data sample set related to the target business; wherein the data samples in the business data sample set include a marking result and a marking reason; the marking reason indicates the reason for generating the marking result; For each first prompt word in the at least one first prompt word, perform the following steps: Obtaining a first evaluation result of the first prompt word on the business data sample set, and determining error sample information corresponding to the first prompt word based on the first evaluation result; the error sample information includes multiple error data samples; Sorting the plurality of erroneous data samples according to the degree of difference between the first evaluation results and the annotation results of the plurality of erroneous data samples, and / or the number of occurrences of the annotation reasons of the plurality of erroneous data samples in the business data sample set, to obtain a sorting result of the plurality of erroneous data samples; In the error sample information corresponding to the first prompt word, the plurality of error sample data are arranged according to the sorting result of the plurality of error data samples; Sending a target prompt word for optimizing the first prompt word to the first language model, wherein the target prompt word includes the first prompt word and error sample information corresponding to the first prompt word; A second prompt word output by the first language model is obtained; the second prompt word is generated by the first language model by optimizing the first prompt word based on the error sample information.

2. The method according to claim 1, characterized in that The error sample information includes: an error data sample, a labeling result of the error data sample, and a first evaluation result of the error data sample.

3. The method according to claim 1, characterized in that The obtaining of a first evaluation result of the first prompt word on the business data sample set includes: Sending a first evaluation prompt word for evaluating the first prompt word to a second language model, where the first evaluation prompt word includes the first prompt word and the business data sample set; A first evaluation result output by the second language model is received.

4. The method according to claim 1, wherein The acquiring of at least one first prompt word associated with the target service includes: Get the initial prompt word; Sending an optimized prompt word for optimizing the initial prompt word to a third language model, wherein the optimized prompt word includes the initial prompt word; Receive at least one first prompt word output by the third language model.

5. The method according to claim 1, wherein The acquiring of at least one first prompt word associated with the target service includes: Obtaining an initial prompt word, and determining at least one target data sample from the business data sample set; The initial prompt word and the at least one target data sample are concatenated to obtain at least one first prompt word.

6. The method according to claim 5, characterized in that The determining at least one target data sample from the business data sample set includes: Determine at least one target data sample according to the number of occurrences of the marking reason in the business data sample set; or, Based on the semantic information of the data samples in the business data sample set, the data samples in the business data sample set are clustered to determine a clustering result; and according to the clustering result, at least one target data sample is determined.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: Determining an optimization effect of each of the second prompt words; A third prompt word is determined from at least one second prompt word according to the optimization effect of each second prompt word.

8. The method according to claim 1, characterized in that The determining of the optimization effect of each second prompt word includes: For each second prompt word, perform the following steps: Sending a second evaluation prompt word for evaluating an optimization effect of the second prompt word to the second language model, where the second evaluation prompt word includes the second prompt word, a business data sample set, and a method for determining the optimization effect; Receive the optimization effect output by the second language model.

9. The method according to claim 7, characterized in that The method further comprises: The third prompt word is determined as the first prompt word in the next round of prompt word optimization process, so as to optimize the third prompt word in the next round of prompt word optimization process.

10. The method according to any one of claims 1 to 9, characterized in that The error sample information includes an error data sample, and the method further includes: Determining a difficult data sample from the erroneous data sample according to a degree of difference between the first evaluation result and the labeling result of the erroneous data sample; The difficult data sample is presented.

11. The method according to any one of claims 1 to 9, characterized in that The first prompt word includes a plurality of prompt words, and the method further includes: determining a prediction result based on first evaluation results of at least part of the first prompt words for the same data sample; presenting the prediction results; Wherein, when the task type indicated by the first prompt word is a classification task, the prediction result is the evaluation result with the largest number of occurrences among the first evaluation results of at least some of the first prompt words for the same data sample; When the task type indicated by the first prompt word is a regression task, the prediction result is an average of first evaluation results of at least some of the first prompt words for the same data sample.

12. A prompt word optimization device, characterized in that: The device comprises: an acquisition module, configured to acquire at least one first prompt word associated with a target business, and acquire a business data sample set related to the target business; wherein the data samples in the business data sample set include a labeling result and a labeling reason; the labeling reason indicates a reason for generating the labeling result; An execution module is configured to perform the following steps for each of the at least one first prompt word: obtaining a first evaluation result of the first prompt word on the business data sample set, and determining error sample information corresponding to the first prompt word based on the first evaluation result; the error sample information including multiple error data samples; sorting the multiple error data samples according to a degree of difference between the first evaluation results and the annotation results of the multiple error data samples and / or a number of occurrences of the annotation reasons of the multiple error data samples in the business data sample set to obtain a sorted result of the multiple error data samples; arranging the multiple error sample data in the error sample information corresponding to the first prompt word according to the sorted result of the multiple error data samples; sending a target prompt word for optimizing the first prompt word to a first language model, the target prompt word including the first prompt word and the error sample information corresponding to the first prompt word; and obtaining a second prompt word output by the first language model; the second prompt word being generated by the first language model by optimizing the first prompt word based on the error sample information.

13. An electronic device, characterized in that: The electronic device includes a processor and a memory; The processor is configured to execute instructions stored in the memory, so that the electronic device performs the method according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that The method comprises instructions for instructing an electronic device to execute the method according to any one of claims 1 to 11.

15. A computer program product, characterized in that The computer program product comprises computer-readable instructions for implementing the method according to any one of claims 1 to 11.

Images