Contract auditing method and device, electronic equipment, storage medium and program product

Abstract

The invention relates to the technical field of computers, and particularly discloses a contract auditing method and device, electronic equipment, a storage medium and a program product, and the method comprises the steps: obtaining a candidate contract text; target key fields are determined according to contract types and contract categories of the candidate contract texts, target contract texts are extracted from the candidate contract texts, and the target contract texts are texts related to the target key fields; inputting the target contract text and the first prompt word into a first model to obtain an audit result output by the first model; the first model is used for auditing the target contract text according to the indication of the first prompt word. On the basis, the contract data can be managed more efficiently, the overall operation efficiency is improved, the management cost is reduced, and meanwhile potential risks caused by human negligence are reduced.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a contract review method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] Currently, both in the aviation industry and other sectors, a large number of contracts are processed annually. These contracts cover various types, including but not limited to commercial contracts, employment contracts, and service contracts. Whenever staff need to retrieve specific clauses or content from a contract, they must first sift through a massive amount of documents to find the relevant contract and carefully review its contents to determine if it contains the required information. This step often consumes a significant amount of time and effort.

[0003] The shortcomings of existing contract management processes are becoming increasingly apparent. On the one hand, the process is extremely time-consuming, requiring staff to spend a significant amount of time sifting through numerous documents to select the necessary contracts. On the other hand, the cumbersome search process leads to overall low efficiency, increasing not only the workload of staff but also causing numerous inconveniences to the relevant systems. This inefficient management approach not only affects work progress but may also lead to the omission or delay of important information, thereby impacting the accuracy and timeliness of decision-making.

[0004] Therefore, how to improve the efficiency of contract review and reduce potential risks caused by human negligence has become an urgent problem to be solved. Summary of the Invention

[0005] This application provides a contract review method, apparatus, electronic device, storage medium, and program product to improve contract review efficiency and reduce potential risks caused by human negligence.

[0006] In a first aspect, this application provides a contract review method, comprising: obtaining candidate contract texts; determining target key fields based on the contract type and contract category of the candidate contract texts; extracting target contract texts from the candidate contract texts, wherein the target contract texts are texts related to the target key fields; inputting the target contract texts and a first prompt word into a first model to obtain the review result output by the first model; the first model is used to review the target contract texts according to the instructions of the first prompt word.

[0007] After obtaining candidate contract texts, this application determines target key fields based on the contract type and category of the candidate contract texts, and extracts the target contract text from the candidate contract texts. The target contract text and the first prompt word are then input into a first model to obtain the review result output by the first model. Based on this, this application can manage contract data more efficiently, improve overall operational efficiency, reduce management costs, and minimize potential risks caused by human error.

[0008] One possible implementation is that the first prompt word includes at least one of the following: a prompt word for indicating compliance verification, a prompt word for indicating risk identification, and a prompt word for indicating content conflict detection.

[0009] Another possible implementation is that, if the first prompt word includes a prompt word used to indicate compliance verification, the first model is used to perform numerical threshold comparison and semantic compliance matching on the target contract text according to the indication of the first prompt word, and generate an audit result based on the numerical threshold comparison result and the semantic compliance matching result; the audit result includes the compliance verification result.

[0010] Another possible implementation, where the first prompt includes a prompt for indicating risk identification, is that the first model is used to identify keyword risks and implicit risks in the target contract text according to the indication of the first prompt, and to generate an audit result based on the keyword risk identification result and the implicit risk identification result; the audit result includes the risk identification result; the keyword risk identification result represents the reasonableness of keywords in the context of the target contract text; the implicit risk identification result represents the potential risks identified based on the relationship between different clauses in the target contract text.

[0011] Another possible implementation, where the first prompt word includes a prompt word for indicating content conflict detection, is that the first model is used to perform content conflict detection on the target contract text according to the indication of the first prompt word, and generate an audit result based on the content conflict detection result; the audit result includes the content conflict detection result; the content conflict detection result includes the conflict detection result for the detection content of different clauses in the target contract text; the detection content includes at least one of time sequence, amount and rights and obligations logic.

[0012] Another possible implementation method, the above "determine the target key fields based on the contract type and contract category of the candidate contract text, and extract the target contract text from the candidate contract text", specifically includes: inputting the candidate contract text and the second prompt word into the second model to obtain the target contract text output by the second model; the second model is used to identify the contract type and contract category of the candidate contract text according to the indication of the second prompt word, and determine the target key fields based on the contract type and contract category, and extract the text related to the target key fields.

[0013] Another possible implementation, the above "obtaining candidate contract text" specifically includes: obtaining the initial contract text; preprocessing the initial contract text to obtain candidate contract text; the preprocessing includes cleaning redundant symbols and uniform text encoding.

[0014] Secondly, this application provides a contract review device, comprising: an acquisition module for acquiring candidate contract texts; a determination module for determining target key fields based on the contract type and contract category of the candidate contract texts, and extracting target contract text from the candidate contract texts, wherein the target contract text is text related to the target key fields; and a processing module for inputting the target contract text and a first prompt word into a first model to obtain the review result output by the first model; wherein the first model is used to review the target contract text according to the indication of the first prompt word.

[0015] Thirdly, this application provides an electronic device comprising: a processor and a memory; the memory storing processor-executable instructions; when the processor is configured to execute the instructions, causing the electronic device to implement the method of the first aspect described above.

[0016] Fourthly, this application provides a computer-readable storage medium comprising: computer software instructions; which, when executed in an electronic device, cause the electronic device to implement the method described in the first aspect.

[0017] Fifthly, this application provides a computer program product comprising a computer program; when the computer program is run in an electronic device, the electronic device performs the method described in the first aspect.

[0018] It should be noted that the aforementioned computer execution instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the electronic device, or it may be packaged separately from the processor of the electronic device; this application does not limit this.

[0019] The beneficial effects of aspects two through five in this application are described in the corresponding manner in aspect one, and will not be repeated hereafter. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating a contract review method provided in an embodiment of this application;

[0022] Figure 2 A schematic diagram of the composition of a contract review device provided in this application embodiment;

[0023] Figure 3This is a schematic diagram of a possible structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0025] The terms "first" and "second," etc., used in the specification and drawings of this application are used to distinguish different objects or to distinguish different treatments of the same object, rather than to describe a specific order of objects.

[0026] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0027] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0028] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.

[0029] In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0030] Currently, both the legal system and other related systems process tens of thousands of contracts annually. These contracts cover various types, including but not limited to commercial contracts, employment contracts, and service contracts. Whenever specific clauses or content within a contract needs to be retrieved, staff must first sift through a massive amount of documents to find the relevant contracts. This step often consumes a significant amount of time and effort, as they need to meticulously sift through mountains of documents. Then, staff must carefully review the specific content of the contract to determine if it contains the required information.

[0031] This process is not only time-consuming and labor-intensive, but also inefficient, causing numerous inconveniences to the work. First, staff need to sift through numerous documents to find the target contract, a process that is inherently tedious and inefficient. Second, even after finding the relevant contract, reviewing its details requires patience and meticulousness, further extending the processing time.

[0032] Given the current situation, the shortcomings of the existing contract management process are becoming increasingly apparent. On the one hand, the contract management process is extremely time-consuming, requiring staff to spend a significant amount of time sifting through numerous documents to select the necessary contracts, which occupies a large portion of their daily work. On the other hand, the cumbersome search process results in low overall work efficiency, which not only increases the workload of staff but also causes numerous inconveniences for the relevant systems. This inefficient management approach not only affects work progress but may also lead to the omission or delay of important information, thereby impacting the accuracy and timeliness of decision-making. Therefore, how to improve contract review efficiency and reduce potential risks caused by human negligence has become an urgent problem to be solved.

[0033] To address the aforementioned technical problems, this application provides a contract review method, apparatus, electronic device, storage medium, and program product. The approach involves: acquiring candidate contract texts; determining target key fields based on the contract type and category of the candidate contract texts; extracting target contract text from the candidate contract texts, where the target contract text is text related to the target key fields; inputting the target contract text and a first prompt word into a first model to obtain the review result output by the first model; and using the first model to review the target contract text according to the instructions of the first prompt word. Based on this, this application can manage contract data more efficiently, improve overall operational efficiency, reduce management costs, and minimize potential risks caused by human error.

[0034] The embodiments provided in this application will now be described in detail with reference to the accompanying drawings.

[0035] See Figure 1 This is a flowchart illustrating a contract review method provided in an embodiment of this application. Figure 1 As shown, the contract review method provided in this application specifically includes the following steps S101-S103.

[0036] S101: Obtain candidate contract text.

[0037] In some embodiments, the contract text content can be extracted by identifying the uploaded contract file and used as candidate contract text.

[0038] For example, contract documents can be uploaded via a World Wide Web interface, mobile device, or application programming interface (API). The contract documents can be in PDF, Word, or image format. Multimodal text parsing technology can be used to extract contract text content from contract documents of different formats.

[0039] One possible implementation involves obtaining candidate contract texts, which may include: obtaining initial contract texts; preprocessing the initial contract texts to obtain candidate contract texts; the preprocessing includes cleaning redundant symbols and uniform text encoding.

[0040] Specifically, the initial contract text can be the uploaded contract text content. The initial contract text is preprocessed, including cleaning redundant symbols and unifying text encoding, to obtain standardized text content, which serves as the candidate contract text.

[0041] Based on this, the contract review method provided in this application can quickly complete the processing of the contract format before the contract is reviewed by preprocessing the initial contract text, thereby improving processing efficiency and reducing labor costs.

[0042] S102: Based on the contract type and contract category of the candidate contract text, determine the target key fields, and extract the target contract text from the candidate contract text. The target contract text is the text related to the target key fields.

[0043] Contract type refers to the category or form of a specific contract. It is a classification of contracts based on certain common characteristics or elements. The classification of contract types helps parties choose applicable legal norms based on the specific characteristics of the contract, and thereby determine their rights and obligations. For example, contract types can include sales contracts, lease contracts, service contracts, etc.

[0044] Contract categories refer to the classification of contracts based on their nature, characteristics, or legal attributes. This classification helps people better understand and identify different types of contracts, thus enabling more accurate application of legal norms. For example, based on whether the contract involves monetary payment, contracts can be categorized into onerous contracts and gratuitous contracts.

[0045] In some embodiments, key fields corresponding to different contract types and categories are pre-configured. After obtaining candidate contract texts, the target key fields are automatically identified from the pre-configured key fields corresponding to different contract types and categories based on the contract type and category of the candidate contract texts, and the target contract text is extracted from the candidate contract texts based on the target key fields.

[0046] One possible implementation is that, after determining the target key field, the paragraph containing the target key field can be used as the target contract text based on its position in the candidate contract text.

[0047] For example, target key fields may include contract amount, term, and liability terms.

[0048] Industry rules may change for different contract types and categories. Therefore, when industry rules change, the key fields corresponding to different contract types and categories that are pre-configured will also change accordingly. This allows for dynamic adjustment of the field extraction strategy to ensure data accuracy.

[0049] S103: Input the target contract text and the first prompt word into the first model to obtain the review result output by the first model; the first model is used to review the target contract text according to the instructions of the first prompt word.

[0050] In some embodiments, the first model may be a large language model. A large language model refers to a deep learning model trained using a large amount of text data, capable of generating natural language text or understanding the meaning of spoken text.

[0051] Specifically, a first prompt word can be input into the first model. The first prompt word can be a single word or a short phrase. The first prompt word is used to instruct the first model to perform the corresponding task according to the prompt word.

[0052] In some embodiments, the target contract text and a first prompt word are input into a first model. The first model can review the target contract text according to the instructions of the first prompt word, and finally obtain the review result output by the first model. The review result is in natural language text form and may include partial text content extracted from the target contract text, as well as review suggestions for the target contract text.

[0053] When generating audit results that include audit recommendations, key supporting evidence, such as legal provisions and historical case numbers, can be marked in the audit recommendations to allow users to trace the source of the recommendations.

[0054] Furthermore, through continuous learning and training, the first model can constantly optimize its performance, improving its understanding of various contract terms and the accuracy of risk assessment. This will further enhance the intelligence level of contract management, making the entire review process smoother and more efficient. The system can also provide users with more intuitive and detailed reference information based on historical data and existing contract information.

[0055] In some embodiments, after obtaining the review results output by the first model, the review results can be displayed on the front-end interface.

[0056] Based on this, the contract review method provided in this application, after obtaining candidate contract texts, determines target key fields according to the contract type and category of the candidate contract texts, extracts the target contract text from the candidate contract texts, and inputs the target contract text and the first prompt word into a first model to obtain the review result output by the first model. Based on this, this application can manage contract data more efficiently, improve overall operational efficiency, reduce management costs, and reduce potential risks caused by human error.

[0057] One possible implementation involves determining the target key fields based on the contract type and contract category of the candidate contract text, and extracting the target contract text from the candidate contract text. Specifically, this may include: inputting the candidate contract text and the second prompt word into a second model to obtain the target contract text output by the second model; the second model is used to identify the contract type and contract category of the candidate contract text according to the indication of the second prompt word, and determine the target key fields based on the contract type and contract category, and extract the text related to the target key fields.

[0058] In some embodiments, the second model can be a Qwen-32B-AWQ vertical domain large-scale model, which, through domain pre-training and fine-tuning, deeply understands the semantics and legal logic of contract terms. The Qwen-32B-AWQ vertical domain large-scale model is an inference model that significantly reduces memory requirements and improves inference efficiency through AWQ (activation-aware weight quantization) quantization technology.

[0059] Specifically, the candidate contract text and the second prompt word are input into the second model to obtain the target contract text output by the second model. The second prompt word instructs the second model to identify the contract type and category of the candidate contract text. After identifying the contract type and category of the candidate contract text, the target key fields corresponding to the candidate contract text are obtained according to the pre-configured key field correspondence. Based on the target key fields, the corresponding text is extracted as the target contract text.

[0060] For example, the target contract text output by the second model is structured contract data, including terms such as payment terms and default terms; and numerical fields such as amount and term.

[0061] Based on this, the contract review method provided in this application, by inputting candidate contract text and second prompt words into a second model to obtain the target contract text output by the second model, can automatically extract structured information from the candidate contract text and generate machine-processable contract data. This improves the accuracy of text extraction and thus enhances the efficiency of contract review.

[0062] In one possible implementation, the first prompt word includes at least one of a prompt word for indicating compliance verification, a prompt word for indicating risk identification, and a prompt word for indicating content conflict detection.

[0063] Specifically, the first prompt word instructs the first model to review the target contract text, and the first prompt word includes review suggestions for the target contract text. The review results can serve as review suggestions to help users better review the contract text.

[0064] Users can input their review requests through natural language interaction or voice recognition, without any restrictions.

[0065] In some embodiments, dynamic knowledge fusion and precise review suggestions can be generated through the deep integration of the Qwen-32B-AWQ vertical domain large model and retrieval-augmented generation (RAG) technology.

[0066] Specifically, RAG technology can dynamically retrieve the latest regulations, industry conventions, and corporate compliance requirements to ensure the timeliness and compliance of recommendations. It also combines this with a historical case database to match similar contract risk scenarios and provide relevant experience for handling such situations.

[0067] Based on this, the contract review method provided in this application includes at least one of the following as the first prompt word: a prompt word for indicating compliance verification, a prompt word for indicating risk identification, and a prompt word for indicating content conflict detection. Therefore, the target contract text can be reviewed based on the content included in the first prompt word, significantly improving the accuracy and efficiency of contract review and reducing operational risks.

[0068] In one possible implementation, if the first prompt word includes a prompt word for indicating compliance verification, the first model is used to perform numerical threshold comparison and semantic compliance matching on the target contract text according to the indication of the first prompt word, and generate an audit result based on the numerical threshold comparison result and the semantic compliance matching result; the audit result includes the compliance verification result.

[0069] Among them, numerical threshold comparison refers to dynamically comparing numerical terms such as contract amount, proportion, and term with current laws and regulations to mark the risk of exceeding the limit.

[0070] Semantic compliance matching refers to vectorizing the text of a clause and calculating its semantic similarity with a regulatory database to identify ambiguous or implicit violations. If the semantic similarity is greater than a preset similarity threshold, the clause is considered compliant; for example, the preset similarity threshold could be 0.9.

[0071] After the first model performs numerical threshold comparison and semantic compliance matching on the target contract text according to the instructions of the first prompt word, the first model can identify the risk of exceeding the numerical threshold and the compliance of the text semantics, and output the review results. The review results include numerical threshold comparison results and semantic compliance matching results in natural language text form, as well as review suggestions based on the numerical threshold comparison results and semantic compliance matching results.

[0072] Based on this, the text review method provided in this application generates review results by comparing numerical thresholds and semantic compliance of the target contract text. This enables dynamic verification of the compliance of contract clauses, significantly improving the accuracy and efficiency of contract review.

[0073] In another possible implementation, where the first prompt word includes a prompt word for indicating risk identification, the first model is used to perform keyword risk identification and implicit risk identification on the target contract text according to the indication of the first prompt word, and generate an audit result based on the keyword risk identification result and the implicit risk identification result; the audit result includes the risk identification result; the keyword risk identification result characterizes the rationality of keywords in the context of the target contract text; the implicit risk identification result characterizes the potential risks identified based on the relationship between different clauses in the target contract text.

[0074] Specifically, keyword risk identification refers to determining the reasonableness of the clause containing high-risk keywords, such as disclaimers or unlimited liability, by examining the context of the high-risk keywords in the target contract text.

[0075] Hidden risk identification refers to identifying potential risks such as the transfer of liability and exemption from obligations by analyzing the relationships between different clauses in the target contract text. For example, subcontracting clauses can be linked to risks related to quality accountability.

[0076] After the first model identifies keyword risks and implicit risks in the target contract text based on the first prompt, it can generate audit results based on these results. The audit results include the keyword risk identification results and implicit risk identification results in natural language text format, along with audit suggestions based on these results.

[0077] Based on this, the contract review method provided in this application identifies keyword risks and hidden risks in the target contract text, and outputs review results based on the results of keyword risk identification and hidden risk identification. This allows for accurate identification of risks in the target contract text and prompts users to review the clauses containing those risks, thus improving the efficiency and security of contract review.

[0078] In another possible implementation, if the first prompt word includes a prompt word for indicating content conflict detection, the first model is used to perform content conflict detection on the target contract text according to the indication of the first prompt word, and generate an audit result based on the content conflict detection result; the audit result includes the content conflict detection result; the content conflict detection result includes the conflict detection result for the detection content of different clauses in the target contract text; the detection content includes at least one of time sequence, amount and rights and obligations logic.

[0079] Specifically, content conflict detection refers to detecting whether there are conflicts in the timing, amount, and logical relationships of different clauses in the target contract text. For example, if the target contract text contains both payment clauses and delivery clauses, and there is a temporal discrepancy between the payment clauses and the delivery clauses, it indicates that there is a content conflict in the target contract text.

[0080] After the first model performs content conflict detection on the target contract text according to the instructions of the first prompt word, the first model can generate an audit result based on the content conflict detection result. The audit result includes the content conflict detection result in natural language text form, as well as audit suggestions based on the content conflict detection result.

[0081] Based on this, the contract review method provided in this application can detect content conflicts in the target contract text and generate review results based on the detection results. This clearly displays the content conflict results of the target contract text, significantly reducing the workload of manual review and providing strong support for subsequent contract management.

[0082] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0083] This application embodiment can divide the contract review device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0084] In some embodiments, this application also provides a contract review apparatus. The contract review apparatus may include one or more functional modules for implementing the contract review method of the above method embodiments.

[0085] For example, Figure 2 This is a schematic diagram illustrating the composition of a contract review device provided in an embodiment of this application. Figure 2 As shown, the contract review device includes: an acquisition module 201, a determination module 202, and a processing module 203.

[0086] The acquisition module 201 is used to acquire candidate contract texts; the determination module 202 is used to determine the target key fields based on the contract type and contract category of the candidate contract texts, and extract the target contract text from the candidate contract texts. The target contract text is the text related to the target key fields; the processing module 203 is used to input the target contract text and the first prompt word into the first model, and obtain the review result output by the first model; the first model is used to review the target contract text according to the instructions of the first prompt word.

[0087] In some embodiments, the first prompt word includes at least one of a prompt word for indicating compliance verification, a prompt word for indicating risk identification, and a prompt word for indicating content conflict detection.

[0088] In other embodiments, the processing module 203 is further configured to, when the first prompt word includes a prompt word for indicating compliance verification, perform numerical threshold comparison and semantic compliance matching on the target contract text according to the indication of the first prompt word, and generate an audit result based on the numerical threshold comparison result and the semantic compliance matching result; the audit result includes the compliance verification result.

[0089] In some other embodiments, the processing module 203 is further configured to, when the first prompt word includes a prompt word for indicating risk identification, perform keyword risk identification and implicit risk identification on the target contract text according to the indication of the first prompt word, and generate an audit result based on the keyword risk identification result and the implicit risk identification result; the audit result includes the risk identification result; the keyword risk identification result characterizes the rationality of keywords in the context of the target contract text; the implicit risk identification result characterizes the potential risks identified based on the relationship between different clauses in the target contract text.

[0090] In some other embodiments, the processing module 203 is further configured to, when the first prompt word includes a prompt word for indicating content conflict detection, perform content conflict detection on the target contract text according to the indication of the first prompt word, and generate an audit result based on the content conflict detection result; the audit result includes the content conflict detection result; the content conflict detection result includes the conflict detection result for the detection content of different clauses in the target contract text; the detection content includes at least one of time sequence, amount and rights and obligations logic.

[0091] In some other embodiments, the determining module 202 is further configured to input the candidate contract text and the second prompt word into the second model to obtain the target contract text output by the second model; the second model is configured to identify the contract type and contract category of the candidate contract text according to the indication of the second prompt word, and determine the target key field according to the contract type and contract category, and extract the text related to the target key field.

[0092] In some other embodiments, the acquisition module 201 is further configured to acquire the initial contract text; preprocess the initial contract text to obtain candidate contract texts; the preprocessing includes cleaning redundant symbols and uniform text encoding.

[0093] In the case of implementing the functions of the integrated modules described above in hardware, this embodiment of the invention provides a possible structural schematic diagram of the electronic device involved in the above embodiments. For example... Figure 3 As shown, the electronic device 300 includes: a processor 302, a communication interface 303, and a bus 304. Optionally, the electronic device 300 may also include a memory 301.

[0094] Processor 302 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 302 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 302 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0095] Communication interface 303 is used to connect to other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0096] The memory 301 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0097] As one possible implementation, the memory 301 can exist independently of the processor 302. The memory 301 can be connected to the processor 302 via a bus 304 and is used to store instructions or program code. When the processor 302 calls and executes the instructions or program code stored in the memory 301, it can implement the contract review method provided in this embodiment of the invention.

[0098] In another possible implementation, the memory 301 can also be integrated with the processor 302.

[0099] Bus 304 can be an extended industry standard architecture (EISA) bus, etc. Bus 304 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0100] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.

[0101] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be executed by computer instructions instructing related hardware. The program can be stored in the aforementioned computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be any of the foregoing embodiments or memory. The aforementioned computer-readable storage medium can also be an external storage device of the aforementioned service invocation device, such as a pluggable hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the aforementioned service invocation device. Further, the aforementioned computer-readable storage medium can include both internal storage units of the aforementioned service invocation device and external storage devices. The aforementioned computer-readable storage medium is used to store the aforementioned computer program and other programs and data required by the aforementioned service invocation device. The aforementioned computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0102] This application also provides a computer program product comprising a computer program that, when run on a computer, causes the computer to perform any of the contract review methods provided in the above embodiments.

[0103] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0104] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0105] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0106] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A contract review method, characterized in that, include: Obtain candidate contract texts; Based on the contract type and contract category of the candidate contract texts, target key fields are determined, and target contract texts are extracted from the candidate contract texts. The target contract texts are texts related to the target key fields. The target contract text and the first prompt word are input into the first model to obtain the review result output by the first model; the first model is used to review the target contract text according to the instructions of the first prompt word.

2. The method according to claim 1, characterized in that, The first prompt word includes at least one of the following: a prompt word for indicating compliance verification, a prompt word for indicating risk identification, and a prompt word for indicating content conflict detection.

3. The method according to claim 2, characterized in that, When the first prompt word includes a prompt word for indicating compliance verification, the first model is used to perform numerical threshold comparison and semantic compliance matching on the target contract text according to the indication of the first prompt word, and generate the audit result based on the numerical threshold comparison result and the semantic compliance matching result; the audit result includes the compliance verification result.

4. The method according to claim 2, characterized in that, When the first prompt word includes a prompt word for indicating risk identification, the first model is used to perform keyword risk identification and implicit risk identification on the target contract text according to the indication of the first prompt word, and generate the audit result based on the keyword risk identification result and the implicit risk identification result; The audit results include risk identification results; The keyword risk identification results characterize the rationality of keywords in the context of the target contract text; The results of the hidden risk identification characterize potential risks identified based on the relationships between different clauses in the target contract text.

5. The method according to claim 2, characterized in that, When the first prompt word includes a prompt word for indicating content conflict detection, the first model is used to perform content conflict detection on the target contract text according to the indication of the first prompt word, and generate the review result based on the content conflict detection result; the review result includes the content conflict detection result; the content conflict detection result includes the conflict detection result for the detected content of different clauses in the target contract text; the detected content includes at least one of time sequence, amount and rights and obligations logic.

6. The method according to claim 1, characterized in that, The step of determining target key fields based on the contract type and contract category of the candidate contract text, and extracting the target contract text from the candidate contract text, includes: The candidate contract text and the second prompt word are input into the second model to obtain the target contract text output by the second model. The second model is used to identify the contract type and contract category of the candidate contract text according to the indication of the second prompt word, and to determine the target key field according to the contract type and the contract category, and extract the text related to the target key field.

7. The method according to claim 1, characterized in that, The process of obtaining candidate contract texts includes: Obtain the initial contract text; The initial contract text is preprocessed to obtain the candidate contract text; the preprocessing includes cleaning redundant symbols and uniform text encoding.

8. A contract review device, characterized in that, include: The acquisition unit is used to acquire candidate contract texts. The determining unit is configured to determine the target key field based on the contract type and contract category of the candidate contract text, and extract the target contract text from the candidate contract text, wherein the target contract text is text related to the target key field; The processing unit is used to input the target contract text and the first prompt word into the first model and obtain the review result output by the first model; The first model is used to review the target contract text according to the instructions of the first prompt word.

9. An electronic device, characterized in that, It includes a processor and a memory, the processor being coupled to the memory; the memory is used to store computer instructions, which are loaded and executed by the processor to enable the computer device to implement the contract review method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes computer-executable instructions that, when executed on a computer, cause the computer to perform the contract review method as described in any one of claims 1 to 7.

11. A computer program product, characterized in that, The computer program product includes a computer program that, when run on an electronic device, causes the electronic device to perform the contract review method as described in any one of claims 1 to 7.

Images