An automatic audit method for engineering projects based on large models

Abstract

The present invention provides an automatic audit method for engineering projects based on a large model. Through an AI large model fine-tuned based on knowledge in the field of engineering auditing, the knowledge rules and audit relationships in the policy documents are automatically extracted according to a preset prompt library. In combination with the audit unit set introduced by the present invention, which includes the audit calculation process and calculation description, the method is understood and used by the AI ​​large model, and a scheduling algorithm model based on the audit calculation unit is generated, thereby obtaining audit doubts of engineering projects and realizing "end-to-end" automated auditing from the "policy document end" to the "audit decision end." In particular, the present invention twice applies the cyclic reverse verification technology of the AI ​​large model to improve the accuracy of automatically generated audit ternary pairs and audit models, thereby improving the reliability of audit results while realizing automated auditing of engineering projects.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence applications, and in particular to an automatic audit method for engineering projects based on a large model. Background Art

[0002] Driven by the digital age and technological advancements, the volume of corporate audit data is growing exponentially. Coupled with intensified market competition and increased compliance requirements, internal audit work at large enterprises is facing severe challenges. Against this backdrop, the transformation to digital, automated auditing is urgent. Automated project auditing is a key application scenario within digital auditing. It aims to automatically and accurately extract audit issues from existing project data, assisting auditors in further investigation, verification, and evidence collection.

[0003] Currently, two representative technologies in the field of digital auditing are knowledge graphs and AI big models. Both have applications in specific business scenarios, including audit knowledge question-and-answering, audit knowledge recommendation, audit anomaly warning, audit report generation, and the construction of intelligent audit systems. However, existing technology approaches cannot fully meet the practical needs of automated auditing of engineering projects. Specifically, while knowledge graphs can query, analyze, and recommend based on clear semantic structures, they lack autonomous learning capabilities, are unable to generate new content, and have limited understanding of natural language text. Furthermore, they lack general applicability, particularly in the field of engineering project auditing. They cannot fully understand the full knowledge contained in policy documents like human auditors do, and often require the collaboration of human auditors to complete specific audit tasks. AI big models, while possessing a rich knowledge base, command-following capabilities, the ability to handle multiple tasks, and automatically generate diverse data, are still black-box and difficult to interpret, requiring verification through manual effort or other systems. Furthermore, because auditing involves massive amounts of data stored in digital platforms, the generation capabilities of AI big models cannot fully integrate with existing audit databases, making it impossible to directly translate the inference results of AI big models into audit decisions. To sum up, the two mainstream technical applications in the current digital auditing field have some drawbacks and cannot truly realize the autonomous audit process from the data segment to the decision-making end. Summary of the Invention

[0004] In order to make up for the deficiencies of the prior art, the present invention provides an automatic audit method for engineering projects based on a large model.

[0005] The present invention is achieved through the following technical solution: an automatic audit method for engineering projects based on a large model, specifically comprising the following steps:

[0006] Step S1: Based on the enterprise's internal digital platform, create an audit unit set S for audit work to extract specific audit data respectively;

[0007] Step S2: Select the document policies used in the project audit as audit knowledge data, perform preprocessing, and use the obtained data set to fine-tune the AI ​​model;

[0008] The document policies used in engineering project audits are selected as audit knowledge data, including four categories: audit policies, laws and regulations, internal corporate regulations, and engineering project audit cases. The collected audit knowledge data is processed through standard preprocessing steps including textualization, cleaning, deduplication, and denoising. The preprocessed text data is converted into a text format suitable for AI large-scale model training to obtain the fine-tuning training dataset T.

[0009] Next, the LORA algorithm is used to fine-tune the large AI model. This algorithm leverages the inherent low-rank nature of the large model and simulates the effect of full parameter fine-tuning by adding low-rank matrices, achieving efficient model adaptive adjustment. This ensures that while retaining general intelligence, it also enhances professionalism in the field of engineering auditing.

[0010] Step S3: Input the text data of the audit policy of the target project as the AI ​​big model. Using the prompt word technology, extract the audit entities and audit relationships to form a candidate audit triple set AT_CAND; obtain the specific audit policy text data of the target project, and perform necessary formatting and preprocessing, including: noise removal, text standardization, word segmentation, and stop word removal. Input the preprocessed audit policy text data into the AI ​​big model; then, using the prompt word technology, let the AI ​​big model automatically extract a series of audit entities and the associated audit relationships;

[0011] Step S4: Send the triples in the obtained candidate audit triple set AT_CAND to the AI ​​big model for reverse verification, remove inaccurate audit triples, and retain the audit triples that have passed the verification to form the set AT_VALID;

[0012] Step S5: Based on the information of each audit unit in the audit unit set S, the audit triples in the set AT_VALID are used to prompt the AI ​​large model to obtain the candidate audit model set AM_CAND;

[0013] Step S6: Send the audit models in the candidate audit model set AM_CAND to the AI ​​big model for reverse verification, remove inaccurate audit models, and retain the audit models that have passed the verification to form the set AM_VALID;

[0014] Step S7, run the audit model code in the audit model set AM_VALID to find the audit doubts of the engineering project; for the audit model set AM_VALID formed in step S6, run the audit model codes it contains one by one on the internal digital platform of the enterprise, and determine the specific audit doubts of the engineering project based on the returned results.

[0015] As a preferred solution, step S1 specifically includes the following steps: for project audit, create an audit unit set S, which includes multiple audit units with parameter lists. Each audit unit is used to automatically extract eight types of original audit data from the database of the digital platform, including funding, progress, contract, quality, personnel information, project information, equipment information, and material. The created audit unit AU i Can be expressed as AU i ={Unit DescriptionAU info , parameter list AU param , executable operation behavior AU opt}, where the audit unit describes AU info Includes description of audit unit function, parameter description, and return type description, which are used to be recognized, understood and used by AI big model; parameter list AU param Used to specify the specific parameters of the operation performed by the audit unit; the executable operation behavior AU opt It is a sequence of operations that can be performed by the audit digital platform to extract specific audit data based on a parameter list.

[0016] As a preferred solution, fine-tuning the AI ​​large model using the LORA algorithm in step S2 specifically includes the following steps:

[0017] Step S2-1: Keep the parameter weight matrix W of the pre-trained AI large model unchanged;

[0018] Step S2-2: Introduce low-rank weight matrix A m,p 、B p,n , where m, p, and n represent the rows of matrix A, the columns of matrix A, and the columns of matrix B, respectively. The product of these two matrices is C m,n It will be used to simulate the parameter weight changes of full parameter fine-tuning; in particular, p is much smaller than m, which means that most of the information in the matrix can be represented by a few principal components, which helps to reduce the complexity of model parameter learning.

[0019] Step S2-3, based on the training data set T, the low-rank weight matrix A is trained using the gradient descent method. m,p 、B p,n After training, the trained weight matrix A is obtained after the number of iterations. * m,p 、B* p,n ;

[0020] Step S2-4, merge weights; * m,p 、B * p,n Multiply them together to get the weight update matrix C of the AI ​​large model parameters * m,n , and update the matrix C * m,n Merge into the parameters of the pre-trained model to obtain the fine-tuned parameter weight matrix W * m,n The specific formula for merging weights is:

[0021] W * m,n =W m,n +C * m,n .

[0022] Furthermore, the number of iterations epoch in step S2-4 is set to 200~500.

[0023] As a preferred solution, step S3 specifically includes the following steps: designing an audit relationship prompt vocabulary P={prompt1, prompt2,...} to prompt the AI ​​large model to automatically identify and extract relevant audit entities and their associated audit relationships to generate structured audit triples.

[0024] As a preferred solution, step S4 specifically includes the following steps: the element AT_CAND in the candidate audit triple set AT_CAND formed in step S3 is i , and then send it to the AI ​​big model for reverse verification through the prompt word technology, asking whether the triple accurately reflects the rule facts in the policy document, and let the big model return a result from the set {accurate, inaccurate}; if it returns "accurate", then AT_CAND i Add the verified triples to the set AT_VALID; if it returns "inaccurate", have it return a similar revised audit triple and verify it again. If the above steps are repeated more than 3 times, discard the triple; otherwise, add the verified triples to the set AT_VALID.

[0025] As a preferred solution, step S5 specifically includes the following steps: sending the information of each audit unit of the accounting unit set S to the AI ​​big model for learning; then, according to the audit triple AT_VALID of the set AT_VALID iPrompt the AI ​​model so that it can reasonably call the audit unit in the set S and verify the result according to the audit triple AT_VALID i Design executable audit code to form a candidate audit model AM_CAND that can detect audit doubts i , and added to the set AM_CAND.

[0026] As a preferred solution, step S6 specifically includes the following steps: the audit model AM_CAND in the candidate audit model set AM_CAND formed in step S5 is i , through the prompt word technology, it is sent to the AI ​​big model for reverse verification, asking whether the audit model accurately reflects the relationship described by the corresponding audit triple, and asking the big model to return a result from the set {accurate, inaccurate}; if it returns "accurate", then AM_CAND i Add the verified audit model set AM_VALID; if it returns "inaccurate", return the revised audit model and verify it again; if the above steps are repeated more than 3 times, discard the audit model; otherwise, add the verified audit model to the set AM_VALID.

[0027] Due to the adoption of the above technical solutions, the present invention has the following beneficial effects compared with the existing technology: (1) The AI ​​big model based on the field of engineering project audit is used to automatically extract audit rule ternary pairs from policy documents, and combined with the audit unit set, an audit model that can be automatically run by the digital platform is automatically generated and used to discover audit doubts, thereby realizing the "end-to-end" automated audit of engineering projects and improving the automation level of audit work. (2) By introducing an audit unit set containing the audit calculation process and calculation description, it can be understood, analyzed and used by the AI ​​big model, so that the generated results have the ability to connect with the existing digital platform, solving the problem that the AI ​​big model cannot connect with the digital audit platform database, and providing reference and reference for the existing automatic audit scheme based on the AI ​​big model and the knowledge graph. (3) In addition, by introducing the cyclic reverse verification step twice in the AI ​​big model generation process, the accuracy of the audit results is improved, the quality of the audit work is guaranteed, and technical support is provided for the actual promotion and application of the present invention. In summary, the method of the present invention has certain promotion and application value in the field of digital audit.

[0028] Additional aspects and advantages of the invention will become apparent from the description which follows, or may be learned by practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments with reference to the following drawings, in which:

[0030] Figure 1 It is the workflow diagram of the present invention. DETAILED DESCRIPTION

[0031] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0032] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0033] The following combination Figure 1 The automatic audit method for engineering projects based on a large model according to an embodiment of the present invention is described in detail.

[0034] like Figure 1 As shown, the present invention proposes an automatic audit method for engineering projects based on a large model, introduces an audit unit set, and realizes the docking of the generative model with the existing digital platform; introduces a two-cycle reverse verification technology in the step to ensure the accuracy of the audit model generated by the AI ​​large model; and realizes the "end-to-end" automatic audit of engineering projects from the "policy document end" to the "audit decision end". Specifically, it includes the following steps:

[0035] Step S1: Based on the enterprise's internal digital platform, create an audit unit set S for audit work to extract specific audit data respectively; the enterprise's internal digital platform refers to an information technology-based software system that integrates data, applications, users, and businesses in the enterprise, and can run the audit model in the present invention and generate results.

[0036] For engineering project audit, an audit unit set S is created, which contains multiple audit units with parameter lists. Each audit unit is used to automatically extract eight types of original audit data from the database of the digital platform, including funding, progress, contract, quality, personnel information, project information, equipment information, and materials. These eight types of data cover the data involved in engineering project audit and can be used as specific audit basis and evidence for the target project. Furthermore, the audit unit AU created i Can be expressed as AU i ={Unit DescriptionAU info , parameter list AU param , executable operation behavior AU opt}, where the audit unit describes AU infoIncludes description of audit unit function, parameter description, and return type description, which are used to be recognized, understood and used by AI big model; parameter list AU param Used to specify the specific parameters of the operation performed by the audit unit; the executable operation behavior AU opt It is a sequence of operations that can be performed by the audit digital platform to extract specific audit data based on a parameter list.

[0037] Step S2: Select the document policies used in the project audit as audit knowledge data, perform preprocessing, and use the obtained data set to fine-tune the AI ​​model;

[0038] The document policies used in engineering project audits are selected as audit knowledge data, including four categories: audit policies, laws and regulations, internal corporate regulations, and engineering project audit cases. The collected audit knowledge data is processed through standard preprocessing steps including textualization, cleaning, deduplication, and denoising. The preprocessed text data is converted into a text format suitable for AI large-scale model training to obtain the fine-tuning training dataset T.

[0039] Next, the LORA algorithm is used to fine-tune the large AI model. This algorithm leverages the inherent low-rank nature of the large model and simulates the effect of full parameter fine-tuning by adding low-rank matrices, achieving efficient model adaptive adjustment. This ensures that while retaining general intelligence, it also enhances professionalism in the field of engineering auditing.

[0040] The specific fine-tuning steps are as follows: Step S2-1, keep the parameter weight matrix W of the pre-trained AI large model unchanged;

[0041] Step S2-2: Introduce low-rank weight matrix A m,p 、B p,n , where m, p, and n represent the rows of matrix A, the columns of matrix A, and the columns of matrix B, respectively. The product of these two matrices is C m,n It will be used to simulate the parameter weight changes of full parameter fine-tuning; in particular, p is much smaller than m, which means that most of the information in the matrix can be represented by a few principal components, which helps to reduce the complexity of model parameter learning.

[0042] Step S2-3, based on the training data set T, the low-rank weight matrix A is trained using the gradient descent method. m,p 、B p,n After training, after the number of iterations (the number of iterations epoch is set to 200~500), the trained weight matrix A is obtained * m,p 、B * p,n ;

[0043] Step S2-4, merge weights;* m,p 、B * p,n Multiply them together to get the weight update matrix C of the AI ​​large model parameters * m,n , and update the matrix C * m,n Merge into the parameters of the pre-trained model to obtain the fine-tuned parameter weight matrix W * m,n The specific formula for merging weights is:

[0044] W * m,n =W m,n +C * m,n

[0045] In the above steps, through the low-rank matrix A m,p 、B p,n , only a small amount of computing resources are needed to fine-tune the weight matrix of the pre-trained AI large model, so that it not only retains the knowledge of the pre-trained model but also can adapt to the needs of the specific tasks faced by the present invention.

[0046] Step S3: Input the text data of the audit policy of the target project as the AI ​​big model. Using the prompt word technology, extract the audit entities and audit relationships to form a candidate audit triple set AT_CAND; obtain the specific audit policy text data of the target project and perform necessary formatting and preprocessing, including: noise removal, text standardization, word segmentation, and stop word removal. Input the preprocessed audit policy text data into the AI ​​big model; then, using the prompt word technology, let the AI ​​big model automatically extract a series of audit entities and the associated audit relationships;

[0047] An audit relationship prompt vocabulary P = {prompt1, prompt2, ...} is designed to prompt the AI ​​model to automatically identify and extract relevant audit entities and their associated audit relationships, generating structured audit triples. The prompt words in the audit relationship prompt vocabulary P are created through repeated practice by auditors, with the goal of fully extracting audit relationships encompassing various audit elements from policy documents. A possible example based on an engineering project is shown in the table below.

[0048]

[0049] Step S4: Send the triples in the obtained candidate audit triple set AT_CAND to the AI ​​big model for reverse verification, remove inaccurate audit triples, and retain the audit triples that have passed the verification to form the set AT_VALID;

[0050] Research shows that large AI models can perform explainable self-verification of their own reasoning conclusions, solving the problem of deviation from the correct thinking process in large model thinking chain technology, thereby improving reasoning performance.

[0051] The element AT_CAND in the candidate audit triple set AT_CAND formed in step S3 is i , and then send it to the AI ​​big model for reverse verification through the prompt word technology, asking whether the triple accurately reflects the rule facts in the policy document, and let the big model return a result from the set {accurate, inaccurate}; if it returns "accurate", then AT_CAND i Add the verified triples to the set AT_VALID; if it returns "inaccurate", have it return a similar revised audit triple and verify it again. If the above steps are repeated more than 3 times, discard the triple; otherwise, add the verified triples to the set AT_VALID.

[0052] Through the reverse verification process of this step, a relatively accurate audit triple set AT_VALID is obtained. The triples in this set provide a reasoning basis for guiding the AI ​​big model to generate a specific audit model.

[0053] Step S5: Based on the information of each audit unit in the audit unit set S, the audit triples in the set AT_VALID are used to prompt the AI ​​large model to obtain the candidate audit model set AM_CAND;

[0054] In order to make the AI ​​big model understand the purpose and scheduling method of the audit unit and use it to extract specific audit data, the information of each audit unit in the audit unit set S is sent to the AI ​​big model for learning; then, according to the audit triple AT_VALID of the set AT_VALID i Prompt the AI ​​model so that it can reasonably call the audit unit in the set S and verify the result according to the audit triple AT_VALID i Design executable audit code to form a candidate audit model AM_CAND that can detect audit doubts i , and added to the set AM_CAND.

[0055] In this step, the audit triple AT_VALID i ={indirect expenses, not exceeding,30%}, for example, the AI ​​model uses the predetermined prompt words "based on the audit triples of the project, using the functions of each audit unit in the audit unit set to generate an executable program", and the generated program logic is as follows: a Calculate the total cost of the project using AU aCalculate the indirect expenses of the project. If the ratio of indirect expenses to total expenses is greater than 30%, an audit doubt will be returned. a Represents the audit unit that calculates the total funds of the target project, which returns the total funds of the specified project from the database; AU b This audit unit represents the calculation of indirect expenses for the target project, returning the indirect expenses for the specified project from the database. By running the aforementioned audit model program on the digital platform, suspicious audit points for the project can be automatically identified.

[0056] Step S6: Send the audit models in the candidate audit model set AM_CAND to the AI ​​big model for reverse verification, remove inaccurate audit models, and retain the audit models that have passed the verification to form the set AM_VALID;

[0057] The audit model AM_CAND in the candidate audit model set AM_CAND formed in step S5 is i , through the prompt word technology, it is sent to the AI ​​big model for reverse verification, asking whether the audit model accurately reflects the relationship described by the corresponding audit triple, and asking the big model to return a result from the set {accurate, inaccurate}; if it returns "accurate", then AM_CAND i Add the verified audit model set AM_VALID; if it returns "inaccurate", return the revised audit model and verify it again; if the above steps are repeated more than 3 times, discard the audit model; otherwise, add the verified audit model to the set AM_VALID.

[0058] Through the reverse verification process of this step, an accurate audit model set AM_VALID is obtained. The audit model codes in this set can help auditors discover audit doubts in engineering projects.

[0059] Step S7, run the audit model code in the audit model set AM_VALID to discover audit doubts of the engineering project; for the audit model set AM_VALID formed in step S6, run the audit model codes it contains one by one on the internal digital platform of the enterprise, determine the specific audit doubts of the engineering project based on the returned results, guide the auditors to pay attention to these specific doubts, conduct in-depth investigation, verification and evidence collection, and realize the "end-to-end" automatic audit method of engineering projects from the "policy document end" to the "audit doubt end".

[0060] In the description of the present invention, the term "plurality" refers to two or more than two. Unless otherwise expressly defined, the orientations or positional relationships indicated by the terms "upper" and "lower" are based on the orientations or positional relationships shown in the accompanying drawings. They are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or components referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they should not be understood as limitations on the present invention. The terms "connect," "install," and "fix" should be understood in a broad sense. For example, "connection" can mean a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] Throughout this specification, terms such as "one embodiment," "some embodiments," and "specific embodiments" mean that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0062] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An automatic audit method for engineering projects based on a large model, characterized in that: The specific steps include: Step S1: Based on the enterprise's internal digital platform, create an audit unit set S for audit work to extract specific audit data respectively; Step S2: Select the document policies used in the project audit as audit knowledge data, perform preprocessing, and use the obtained data set to fine-tune the AI ​​model; The document policies used in engineering project audits are selected as audit knowledge data, including four categories: audit policies, laws and regulations, internal corporate regulations, and engineering project audit cases. The collected audit knowledge data is processed through standard preprocessing steps including textualization, cleaning, deduplication, and denoising. The preprocessed text data is converted into a text format suitable for AI large-scale model training to obtain the fine-tuning training dataset T. Next, the LORA algorithm is used to fine-tune the large AI model. This algorithm leverages the inherent low-rank nature of the large model and simulates the effect of full parameter fine-tuning by adding low-rank matrices, achieving efficient model adaptive adjustment. This ensures that while retaining general intelligence, it also enhances professionalism in the field of engineering auditing. Step S3: The text data of the audit policy of the target project is used as the input of the AI ​​large model. The audit entities and audit relations are extracted using the prompt word technology to form a candidate audit triple set AT_CAND; Obtain the specific audit policy text data of the target project and perform formatting and preprocessing, including noise removal, text standardization, word segmentation, and stop word removal. Input the preprocessed audit policy text data into the AI ​​big model. Then, using prompt word technology, the AI ​​big model automatically extracts a series of audit entities and their associated audit relationships. Step S4: Send the triples in the obtained candidate audit triple set AT_CAND to the AI ​​big model for reverse verification, remove inaccurate audit triples, and retain the audit triples that have passed the verification to form the set AT_VALID; Step S5: Based on the information of each audit unit in the audit unit set S, the audit triples in the set AT_VALID are used to prompt the AI ​​large model to obtain the candidate audit model set AM_CAND; Step S6: Send the audit models in the candidate audit model set AM_CAND to the AI ​​big model for reverse verification, remove inaccurate audit models, and retain the audit models that have passed the verification to form the set AM_VALID; Step S7, run the audit model code in the audit model set AM_VALID to find the audit doubts of the engineering project; for the audit model set AM_VALID formed in step S6, run the audit model codes it contains one by one on the internal digital platform of the enterprise, and determine the specific audit doubts of the engineering project based on the returned results.

2. The automatic audit method for engineering projects based on a large model according to claim 1 is characterized in that: The step S1 specifically includes the following steps: for the project audit, an audit unit set S is created, which includes multiple audit units with parameter lists, each audit unit is used to automatically extract eight types of original audit data from the database of the digital platform, including funding, progress, contract, quality, personnel information, project information, equipment information, and material. The created audit unit AU i Can be expressed as AU i ={Unit DescriptionAU info , parameter list AU param , executable operation behavior AU opt }, where the audit unit describes AU info Includes description of audit unit function, parameter description, and return type description, which are used to be recognized, understood and used by AI big model; parameter list AU param Used to specify the specific parameters of the operation performed by the audit unit; the executable operation behavior AU opt It is a sequence of operations that can be performed by the audit digital platform to extract specific audit data based on a parameter list.

3. The automatic audit method for engineering projects based on a large model according to claim 1 is characterized in that: In step S2, fine-tuning the AI ​​large model using the LORA algorithm specifically includes the following steps: Step S2-1: Keep the parameter weight matrix W of the pre-trained AI large model unchanged; Step S2-2: Introduce low-rank weight matrix A m,p 、B p,n , where m, p, and n represent the rows of matrix A, the columns of matrix A, and the columns of matrix B, respectively. The product of these two matrices is C m,n It will be used to simulate the parameter weight changes of full parameter fine-tuning; in particular, p is much smaller than m, which means that most of the information in the matrix can be represented by a few principal components, which helps to reduce the complexity of model parameter learning; Step S2-3, based on the training data set T, the low-rank weight matrix A is trained using the gradient descent method. m,p 、B p,n After training, the trained weight matrix A is obtained after the number of iterations. * m,p 、B * p,n ; Step S2-4, merge weights; * m,p 、B * p,n Multiply them together to get the weight update matrix C of the AI ​​large model parameters * m,n , and update the matrix C * m,n Merge into the parameters of the pre-trained model to obtain the fine-tuned parameter weight matrix W * m,n The specific formula for merging weights is:

4. The automatic audit method for engineering projects based on a large model according to claim 2 is characterized in that: The number of iterations epoch in step S2-4 is set to 200-500.

5. The automatic audit method for engineering projects based on a large model according to claim 1 is characterized in that: The step S3 specifically includes the following steps: designing an audit relationship prompt vocabulary P={prompt1, prompt2,...} to prompt the AI ​​large model to automatically identify and extract relevant audit entities and their associated audit relationships to generate structured audit triples.

6. The automatic audit method for engineering projects based on a large model according to claim 1 is characterized in that: The step S4 specifically includes the following steps: the element AT_CAND in the candidate audit triple set AT_CAND formed in step S3 i , and then send it to the AI ​​big model for reverse verification through the prompt word technology, asking whether the triple accurately reflects the rule facts in the policy document, and let the big model return a result from the set {accurate, inaccurate}; if it returns "accurate", then AT_CAND i Add the verified triples to the set AT_VALID; if it returns "inaccurate", let it return a similar revised audit triple and verify it again; if the above steps are repeated more than 3 times, discard the triple; otherwise, add the verified triples to the set AT_VALID.

7. The automatic audit method for engineering projects based on a large model according to claim 1 is characterized in that: The step S5 specifically includes the following steps: sending the information of each audit unit of the accounting unit set S to the AI ​​big model for learning; then, according to the audit triple AT_VALID of the set AT_VALID i Prompt the AI ​​model so that it can reasonably call the audit unit in the set S and verify the result according to the audit triple AT_VALID i Design executable audit code to form a candidate audit model AM_CAND that can detect audit doubts i , and added to the set AM_CAND.

8. The automatic audit method for engineering projects based on a large model according to claim 1 is characterized in that: The step S6 specifically includes the following steps: i , through the prompt word technology, it is sent to the AI ​​big model for reverse verification, asking whether the audit model accurately reflects the relationship described by the corresponding audit triple, and asking the big model to return a result from the set {accurate, inaccurate}; if it returns "accurate", then AM_CAND i Add the verified audit model set AM_VALID; if it returns "inaccurate", return the revised audit model and verify it again; if the above steps are repeated more than 3 times, discard the audit model; otherwise, add the verified audit model to the set AM_VALID.

Images