Contract processing method and device based on intelligent agent

Abstract

The embodiment of the invention provides an agent-based contract processing method and device, and the method comprises the steps: firstly obtaining a multi-modal contract file uploaded at a contract processing node of credit investigation service in a contract processing process, the method comprises the following steps: performing content analysis on a multi-modal contract file through a large language model integrated by an intelligent agent to obtain contract analysis data of a contract processing node, calling a corresponding contract processing tool in a tool set of the intelligent agent, performing contract data processing on the contract analysis data to obtain contract data so as to generate a service calling protocol of credit investigation service, the credit investigation service agreement is generated based on the contract file by means of the intelligent agent.

Description

Technical Field

[0001] This document relates to the field of data processing technology, and in particular to a contract processing method and apparatus based on intelligent agents. Background Technology

[0002] With the continuous development of information technology, the application of credit reporting services is becoming increasingly widespread, and more and more Internet platforms and data service agencies are beginning to build and access credit reporting services. Against this backdrop, the cooperation model of credit reporting services is gradually becoming more popular, and service providers of credit reporting services need to sign a large number of service contracts with various institutional clients. As the scale of services continues to expand, the number of contracts is showing a trend of large-scale growth. In this situation, how to improve the accuracy of contract data and the efficiency of contract processing has become a key focus for all service providers. Summary of the Invention

[0003] This specification provides one or more embodiments of a contract processing method based on an intelligent agent, comprising: acquiring a multimodal contract file uploaded to a contract processing node of a credit reporting service; parsing the content of the multimodal contract file using a large language model integrated by the intelligent agent to obtain contract parsing data from the contract processing node; and invoking a corresponding contract processing tool in the intelligent agent's toolset to process the contract parsing data to obtain contract data, thereby generating a service invocation protocol for the credit reporting service.

[0004] This specification provides one or more embodiments of an agent-based contract processing apparatus, comprising: a file acquisition module configured to acquire a multimodal contract file uploaded to a contract processing node of a credit reporting service; a content parsing module configured to parse the multimodal contract file using a large language model integrated by the agent to obtain contract parsing data from the contract processing node; and a protocol generation module configured to invoke a corresponding contract processing tool within the agent's toolset to process the contract parsing data and obtain contract data, thereby generating a service invocation protocol for the credit reporting service.

[0005] This specification provides one or more embodiments of an agent-based contract processing device, comprising: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to: acquire a multimodal contract file uploaded to a contract processing node of a credit reporting service; parse the content of the multimodal contract file using a large language model integrated by the agent to obtain contract parsing data from the contract processing node; and invoke a corresponding contract processing tool within the agent's toolset to process the contract parsing data to obtain contract data, thereby generating a service invocation protocol for the credit reporting service.

[0006] This specification provides one or more embodiments of a computer-readable storage medium for storing computer-executable instructions. When executed, these instructions implement the following process: acquiring a multimodal contract file uploaded to a contract processing node of a credit reporting service; parsing the multimodal contract file using a large language model integrated by an intelligent agent to obtain contract parsing data from the contract processing node; and invoking a corresponding contract processing tool within the intelligent agent's toolset to process the contract parsing data and obtain contract data to generate a service invocation protocol for the credit reporting service. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in one or more embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 A schematic diagram illustrating the implementation environment of an agent-based contract processing method provided in one or more embodiments of this specification; Figure 2 A flowchart illustrating a contract processing method based on an intelligent agent, provided for one or more embodiments of this specification; Figure 3 A flowchart illustrating a contract processing method based on an intelligent agent, applied to a first credit reporting service scenario, provided for one or more embodiments of this specification. Figure 4 A flowchart illustrating a contract processing method based on an intelligent agent, applied to a second credit reporting service scenario, provided for one or more embodiments of this specification. Figure 5 A schematic diagram of an embodiment of a contract processing device based on an intelligent agent, provided for one or more embodiments of this specification; Figure 6 This is a schematic diagram of the structure of a contract processing device based on an intelligent agent, provided for one or more embodiments of this specification. Detailed Implementation

[0008] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0009] The agent-based contract processing method provided in one or more embodiments of this specification can be applied to the credit reporting service implementation environment. (Refer to...) Figure 1 The implementation environment includes at least: Server 101, Agent 102; Among them, server 101 is used to obtain the multimodal contract file of credit reporting service, call agent 102 to perform content parsing and contract data processing on the multimodal contract file, and generate the service call protocol of credit reporting service based on the contract data returned by agent 102; server 101 can be a single server, a server cluster consisting of several servers, or one or more cloud servers in a cloud computing platform. The intelligent agent 102 is used to input multimodal contract documents into a large language model to generate contract parsing data, and to call contract processing tools to process the contract parsing data to obtain contract data. The intelligent agent 102 can be configured with or connected to a large language model to perform content parsing to obtain contract parsing data. The intelligent agent 102 can also be configured with a toolset to process contract data through the contract processing tools in the toolset. The intelligent agent 102 can be deployed independently, and the server 101 interacts with the intelligent agent 102 through invocation. The intelligent agent 102 can be deployed on a single server, in a server cluster consisting of several servers, or in one or more cloud servers in a cloud computing platform. In addition, the intelligent agent 102 can also be deployed inside the server 101.

[0010] The implementation environment may also include an access terminal 103; the access terminal 103 is used to access credit reporting services, send multimodal contract documents to the server 101 and receive contract parsing data or contract data dialogue data sent by the server 101, and can also display dialogue data; the access terminal 103 may specifically be a mobile phone, personal computer, tablet computer, e-book reader, VR (Virtual Reality) based information interaction device, vehicle terminal, IoT device, wearable smart device, laptop computer and desktop computer, etc.; In this implementation environment, after server 101 obtains the multimodal contract file uploaded to the contract processing node of the credit reporting service, server 101 calls agent 102 to process the multimodal contract file. In response to the call, agent 102 inputs the multimodal contract file into a large language model for content parsing to obtain the contract parsing data of the contract processing node. Agent 102 also calls the corresponding contract processing tool in the toolset to process the contract parsing data and obtain the contract data. Agent 102 returns the obtained contract data to server 101. Server 101 generates the service call protocol for the credit reporting service based on the contract data. In this way, the credit reporting service protocol is generated by calling the agent during the contract processing process.

[0011] It should be noted that, considering that the multimodal contract documents, credit reporting product data, form data, and other related data involved in this specification may, to some extent, constitute the privacy of the service provider or the service client, authorization from the service provider or the service client should be obtained before collecting or transmitting such data to ensure that the data collection or transmission operation complies with relevant data management regulations. For example, data authorization can be granted by relevant personnel of the service provider or the service client when the target application is launched. Specific methods of data authorization include sending a data authorization reminder to relevant personnel of the service provider or the service client, who can then obtain data authorization by confirming the reminder; alternatively, data authorization can be obtained by signing a data authorization agreement. This embodiment does not limit the scope of authorization.

[0012] This specification provides one or more embodiments of a contract processing method based on intelligent agents, as follows: Reference Figure 2 The contract processing method based on intelligent agents provided in this embodiment specifically includes steps S202 to S206.

[0013] Step S202: Obtain the multimodal contract file uploaded at the contract processing node of the credit reporting service.

[0014] In this embodiment, the contract processing node refers to the node used to perform contract processing tasks in the credit reporting service process; the contract processing node may include a credit reporting product selection node, a contract price generation node, and / or a credit reporting service signing node; wherein, the credit reporting product selection node refers to the node for selecting and confirming credit reporting products or signing schemes; for example, after the service provider and the signing party communicate offline and determine the selected credit reporting product or signing scheme, the node configures and confirms the selected credit reporting product or signing scheme online; The contract price generation node refers to the node where the contract price in the contract document is configured based on the confirmation of the credit reporting product or contract plan. For example, it may involve determining the contract price, pricing method, and settlement cycle in the contract document; or it may be the node where the service provider and the contracting party configure the price online after they have agreed on the price of the credit reporting product or contract plan offline. The credit reporting service signing node refers to the online synchronization of legally binding contract documents that have been signed offline. For example, after the service provider and the contracting party have completed the signing of the contract documents offline, when the service provider needs to access and use the credit reporting service, it can enter the credit reporting service signing node to sign the service online, thereby enabling the credit reporting service after completing the online service signing.

[0015] In practice, during the contract processing of credit reporting services, the credit reporting platform (the server of the credit reporting platform) can obtain multimodal contract documents uploaded by the operators of the credit reporting services at the contract processing node. Specifically, the contract documents obtained during the contract processing of credit reporting services can be multimodal contract documents or single-modal contract documents.

[0016] Optionally, the multimodal contract documents include: contract documents obtained by the credit reporting platform and the service client through signing at least one signing scheme or at least one credit reporting contract; the signing method of signing the signing scheme or credit reporting contract may include offline signing and / or online signing.

[0017] The multimodal contract document refers to a contract document that includes multiple data modalities; a multimodal contract document may include text-based contract documents, image-based contract documents, and / or tabular contract documents, and may also include contract documents with other data modalities, such as hybrid contract documents containing mixed layouts of images and text.

[0018] In practical applications, in order to improve the overall efficiency of contract processing and enable the compliance and accuracy of contract documents to be discovered and corrected during the contract signing process, thereby reducing the cost of post-event correction, in this embodiment, the operation of the contract processing node can be carried out simultaneously with the offline contract signing process. Based on this, the contract processing process provided in this embodiment does not need to wait until the offline contract signing process is completed before starting. Instead, the contract processing node can be executed in parallel during the offline contract signing process, such as after the signing plan is determined or after the initial draft of the contract document is determined.

[0019] Step S204: The multimodal contract file is parsed using a large language model integrated by the intelligent agent to obtain the contract parsing data of the contract processing node.

[0020] The intelligent agent refers to an entity that can autonomously perform tasks, make decisions, and learn and adjust according to changes in the environment; optionally, the intelligent agent includes a large language model and a toolset. The Large Language Model (LLM) can be a pre-trained natural language model. It can employ foundation models or pre-trained models. Specifically, the architecture of the Large Language Model can be a neural network architecture with a large number of parameters, a Transform architecture, or other architectures. During execution, the Large Language Model can directly use foundation models or pre-trained models, or it can be fine-tuned based on these models to obtain a Large Language Model for parsing contract documents. The toolset refers to a collection of preset functional modules that can be invoked by an intelligent agent; specifically, the toolset can be a collection of contract processing tools that can be invoked by an intelligent agent for contract processing; the toolset may include verification tools and / or risk assessment tools, and may also include other contract processing tools, such as parameter modification tools for modifying or adjusting specific fields in contract-related data, dialogue generation tools for dialogue generation, and conversion tools for file type conversion.

[0021] In specific implementation, based on the acquisition of multimodal contract files as described above, the acquired multimodal contract files can be parsed by an intelligent agent integrating a large language model. Specifically, the intelligent agent can be invoked to process the multimodal contract files. The intelligent agent responds to the call by receiving the multimodal contract files from the contract processing node and inputs the multimodal contract files into the large language model for content parsing. After obtaining the contract parsing data output by the large language model, the intelligent agent returns the contract parsing data to the server. Here, the contract parsing data from the contract processing node is obtained. Furthermore, during contract processing, besides the aforementioned execution entity being the credit reporting platform's server, which obtains the contract file and calls an intelligent agent to process it, the execution entity can also be an intelligent agent. In this case, when the execution entity is an intelligent agent, the aforementioned acquisition of the multimodal contract file uploaded to the contract processing node of the credit reporting service can be performed by the data acquisition module within the intelligent agent. Further, based on the acquired contract file, the intelligent agent calls a large language model to parse the contract file content to obtain contract parsing data. Subsequently, the intelligent agent can also call its toolset to process the contract parsing data to obtain contract data, and further generate the service call protocol for the credit reporting service through the protocol generation module within the intelligent agent.

[0022] In the specific execution process, during the content parsing of multimodal contract documents, in order to improve the accuracy and efficiency of contract parsing, the contract signing objects and / or signing schemes in the contract documents can be determined and extracted based on the semantic recognition results of semantic recognition of multimodal contract documents; in one optional implementation method provided in this embodiment, the content parsing of multimodal contract documents includes: Perform semantic recognition on multimodal contract documents, and determine the contract signing parties based on the semantic recognition results; Based on the semantic recognition results, the signing scheme of the credit reporting product is extracted, and the credit reporting product data contained in the signing scheme is extracted.

[0023] The contract signing parties refer to the legal entities participating in the signing of credit reporting service contracts. The contract signing parties may include credit reporting agencies, such as institutions that provide credit reporting services, financial institutions, such as institutions that need to sign credit reporting service agreements to access credit reporting services, and data source parties, such as related parties that provide basic data or technical support to credit reporting agencies.

[0024] Specifically, in the process of parsing multimodal contract documents, the large language model first performs semantic recognition on the multimodal contract documents. Based on the semantic recognition results, on the one hand, it can determine the identity attributes of the contract signing parties to identify the contract signing objects; on the other hand, it can also extract the credit reporting product signing scheme based on the semantic recognition results, and further extract the credit reporting product data contained in the signing scheme based on the obtained signing scheme.

[0025] Here, in the process of parsing the content of multimodal contract documents, the above-mentioned methods for determining the contract signing object and extracting credit information product data can all be executed, or one of them can be selected for execution; for example, parsing the content of multimodal contract documents also includes: performing semantic recognition on the multimodal contract documents, and determining the contract signing object based on the semantic recognition results; another example is that parsing the content of multimodal contract documents also includes: extracting the credit information product signing scheme based on the semantic recognition results of the multimodal contract documents, and extracting the credit information product data of the credit information product included in the signing scheme.

[0026] In the specific implementation process, during the content parsing of multimodal contract documents, in order to improve the accuracy and efficiency of pricing configuration in credit reporting services, a large language model can be used to match price templates based on text content and generate forms according to the price templates. In one optional implementation method provided in this embodiment, the content parsing of multimodal contract documents includes: Identify credit reporting products and / or contractual schemes in multimodal contract documents, and extract price configuration data according to the price templates corresponding to the credit reporting products and / or contractual schemes.

[0027] Specifically, at least one price template can be pre-set in the large language model. After the multimodal contract document is input into the large language model, the large language model identifies the credit products and / or the overall contract scheme involved in the multimodal contract document, and matches the pre-set price template based on the identification results. Then, it extracts the price data that matches the price template from the multimodal contract document and fills the price data into the corresponding fields of the price template to obtain the price configuration data.

[0028] For example, after identifying the credit reporting products in the multimodal contract documents and extracting the price data according to the corresponding price template of the credit reporting products, the obtained price configuration data may include single billing type, profit sharing billing type or monthly full progressive billing type. The price configuration data may also include the credit reporting product rate and / or billing cycle.

[0029] Furthermore, during the content parsing of multimodal contract documents, contract term data and / or service call parameters can be extracted from the multimodal contract documents, further providing more accurate contract data for subsequent contract processing. In one optional implementation of this embodiment, the content parsing of multimodal contract documents also includes: Contract term data is obtained by extracting the contract term from the semantic recognition results of multimodal contract documents; The service call parameters of the service client are obtained by parsing the service call parameters of the multimodal contract file.

[0030] The contract term data refers to data used to characterize the effective time range of the contract document. The contract term data may include the start time, end time and / or duration of the contract document. For example, the contract term data may be contract renewal information or contract effectiveness information.

[0031] Optionally, service call parameters may include service response time, data update parameters, and / or service priority. These service call parameters refer to the service quality parameters and / or service indicator parameters of the credit reporting service provided by the service provider to the service client, as stipulated in the contract document. For example, service call parameters may be SLA (Service Level Agreement) clauses, which may include key indicators such as service response time and data update frequency. Furthermore, service call parameters may also include other service call-related parameters, such as interface call frequency thresholds and data retention periods.

[0032] Specifically, during the process of extracting the contract term, the large language model can extract contract term-related fields from the semantic recognition results to obtain contract term data; during the process of parsing service call parameters, the large language model can perform keyword recognition based on the semantic recognition results to obtain the service response time, data update parameters and / or service priority of the service customer, thereby obtaining the service call parameters.

[0033] Similarly, in the process of parsing the content of a multimodal contract file, the above-mentioned implementation methods for extracting the contract term and parsing service call parameters can all be executed, or one of them can be selected for execution. For example, parsing the content of a multimodal contract file also includes: extracting the contract term data from the semantic recognition results of the multimodal contract file; or, for another example, parsing the content of a multimodal contract file also includes: parsing the service call parameters of the multimodal contract file to obtain the service call parameters of the service client; furthermore, in the process of parsing the service call parameters, the parsed service call parameters can be at least one of the following: service response time, data update parameters, and service priority, or they can be other service call parameters, or they can be a combination of at least one of the above service call parameters and other service call parameters.

[0034] It should be noted that the above-described implementation methods for parsing multimodal contract documents can be combined in any form according to actual execution needs, or can be adapted, modified, changed, or deleted and then combined in any form according to actual execution needs. For example, parsing multimodal contract documents includes: performing semantic recognition on the multimodal contract documents, determining the contract signing object based on the semantic recognition results; identifying credit reporting products in the multimodal contract documents, extracting price configuration data according to the price template corresponding to the credit reporting products, and parsing service call parameters in the multimodal contract documents to obtain service call parameters for the service customers. Alternatively, content parsing of multimodal contract documents can be performed, including: extracting the signing scheme of credit reporting products based on the semantic recognition results of the multimodal contract documents, extracting the credit reporting product data contained in the signing scheme, and extracting price data according to the price template corresponding to the signing scheme to obtain price configuration data; wherein, the contract parsing data can be at least one of price configuration data, contract term data, and service call parameters, or it can be the contract signing object and / or credit reporting product data provided above, or it can be contract parsing data composed of at least one of price configuration data, contract term data, and service call parameters and data other than these three. The process of obtaining contract parsing data here is similar to the process of obtaining contract parsing data described above, and will not be repeated here.

[0035] Furthermore, it should be noted that the above-described implementation method for parsing multimodal contract documents can be combined in any form according to the needs of the contract processing nodes during actual execution. Alternatively, it can be adapted, modified, or deleted according to the needs of the contract processing nodes during actual execution before being combined in any form. For example, parsing multimodal contract documents includes: extracting the signing scheme of credit reporting products from the semantic recognition results of the multimodal contract documents of the first contract processing node, and extracting the credit reporting product data of the credit reporting products included in the signing scheme; identifying the credit reporting products and signing schemes in the multimodal contract documents of the second contract processing node, and extracting price data according to the price templates corresponding to the credit reporting products and signing schemes to obtain price configuration data; performing semantic recognition on the multimodal contract documents of the third contract processing node, determining the contract signing object based on the semantic recognition results, and parsing the service call parameters of the service customer from the multimodal contract documents of the third contract processing node.

[0036] Step S206: The corresponding contract processing tool is invoked in the toolset of the intelligent agent to process the contract parsing data and obtain contract data, so as to generate the service invocation protocol of the credit reporting service.

[0037] In specific implementation, based on the contract parsing data output by the large language model obtained above, an intelligent agent can be further invoked to process the contract parsing data. Specifically, during the process of invoking the intelligent agent to process the contract parsing data, the intelligent agent responds to the call and determines the corresponding contract processing tool in the toolset. The invoked contract processing tool processes the contract parsing data to obtain the contract data, and the intelligent agent receives and returns the contract data to the server. Here, a service invocation protocol for obtaining the contract data and generating credit reporting services based on the contract data is established.

[0038] As described above, the toolset refers to a collection of preset functional modules that can be invoked by intelligent agents; specifically, the toolset can be a collection of contract processing tools that can be invoked by intelligent agents for contract processing; the contract processing tool refers to a functional unit in the toolset used to process contract data, and the contract processing tool can be a preset functional module, functional component, or callable interface; The toolset may include validation tools and / or risk assessment tools. In addition, it may include other contract processing tools, such as parameter modification tools for modifying or adjusting specific fields in contract-related data, dialogue generation tools for dialogue generation, and conversion tools for file type conversion.

[0039] In this process, by introducing contract processing tools and processing contract data based on these tools, the accuracy and efficiency of contract data processing can be improved. Therefore, the contract processing tool can be determined through contract processing nodes. In one optional implementation of this embodiment, the contract processing tool is determined in the following manner: Based on the contract processing node, determine the corresponding contract processing tool in the toolset for that contract processing node.

[0040] Specifically, during the process of responding to a call to process contract data, the intelligent agent can determine the contract processing tool corresponding to the currently executing contract processing node from the toolset. Subsequently, the intelligent agent can call the determined contract processing tool to process the contract data.

[0041] In addition to determining the contract processing tool based on the contract processing node, the contract processing tool can also be determined based on the multimodal contract document. In another optional implementation provided in this embodiment, the contract processing tool is determined in the following way: Identify the file types of multimodal contract documents and / or the contract processing tools corresponding to the contract parsing data in the toolset.

[0042] Specifically, during the process of intelligent agents responding to calls to process contract data, they can analyze and determine the file type of multimodal contract files, such as text files, image files, or table files. They can also analyze key fields in the acquired contract parsing data and determine the contract processing tool based on the file type or contract parsing data; or, they can determine the contract processing tool based on both the file type and contract parsing data.

[0043] In the specific execution process, during the contract data processing of the parsed contract data, to improve processing efficiency, different configuration forms can be configured for different contract processing nodes, and the form data can be automatically filled based on the configuration forms. Simultaneously, to reduce the possibility of erroneous operations during contract processing, confidence level detection can be performed on the parsed contract data, and confidence level prompts can be provided, so that operators can pay attention to and verify low-confidence data. In an optional implementation method provided in this embodiment, contract data processing is implemented in the following way: Extract form data from the contract parsing data based on the configuration form corresponding to the contract processing node and populate it into the configuration form to obtain the configuration form data of the contract processing node; The confidence level of the contract parsing data output by the large language model is displayed on the interactive page.

[0044] The confidence level indicator refers to a visual identifier used to characterize the level of data credibility. Specifically, after a quantitative assessment of the credibility of contract parsing data, a visual identifier displayed on the interactive page indicates the level of credibility of the contract parsing data to the operations staff. Different confidence level indicators represent different levels of confidence. For example, confidence levels can be divided into low confidence, medium confidence, and high confidence based on the color rendering method of the confidence level indicator. For instance, a medium confidence level indicator can be displayed in the related area of ​​contract term data or price configuration data.

[0045] Specifically, during contract data processing, the system can match the configuration form corresponding to the current contract processing node, identify and extract data information matching the form fields in the configuration form from the contract parsing data, and populate the data to obtain the configuration form data for the contract processing node. Simultaneously, it can output the confidence level after probabilistically evaluating the accuracy of the contract parsing data based on a large language model, and provide visual confidence level prompts for relevant fields in the contract parsing data on the interactive page.

[0046] In practical applications, when parsing multimodal contract documents using a large language model, there may be issues such as significant errors in the contract parsing data output by the large language model due to text ambiguity, formatting issues, or semantic ambiguity in the multimodal contract documents. To address this, verification tools and / or risk assessment tools can be introduced. Verification tools can be used to check the consistency between the parsed contract data and the original contract documents, and / or risk assessment tools can be used to rate the risks of the parsed contract data and generate risk warnings, thereby ensuring data accuracy and improving the reliability of contract processing. In one optional implementation of this embodiment, the corresponding contract processing tool is invoked to process the contract parsing data to obtain contract data, including: Use the verification tool to verify the consistency between the multimodal contract file and the price configuration data, contract term data, and / or service call parameters, and obtain the verification results; Use risk assessment tools to perform risk rating on price configuration data, contract term data, and / or service call parameters, and display risk rating reminders on the interactive page.

[0047] Specifically, in the process of processing contract parsing data, on the one hand, a verification tool can be called to verify the consistency between the original data in the multimodal contract file and the contract parsing data output by the large language model. Specifically, the original data is compared with the price configuration data, contract term data, and / or service call parameters, and the verification results are obtained. On the other hand, a risk assessment tool can be called to perform risk rating on the price configuration data, contract term data, and / or service call parameters based on preset risk rules, and generate risk warnings. These risk warnings can then be displayed on the interactive page through message reminders, pop-ups, or icon rendering, thereby providing risk rating reminders.

[0048] For example, after calling the verification tool to verify the consistency between the multimodal contract document and the price configuration data, the verification result may be: there is a semantic difference between "50 yuan / transaction" in the multimodal contract document and "50 yuan / time" in the price configuration data; based on this, operations personnel can confirm and / or adjust according to the verification result; as another example, after calling the risk assessment tool to perform risk rating on the contract term data, a risk warning or risk report of "contract renewal period does not match the content of the multimodal contract document" can be generated and displayed on the interactive page to remind operations personnel of the risk rating, so that operations personnel can make secondary confirmation through the interactive page.

[0049] In this process, during the risk rating of price configuration data, contract term data, and / or service call parameters, price deviation threshold verification, term deviation threshold verification, and / or parameter deviation threshold verification can be performed based on preset rules. In one optional implementation of this embodiment, risk rating of price configuration data, contract term data, and / or service call parameters includes: Read the deviation verification rules corresponding to the price configuration data, contract term data, and / or service call parameters; The price configuration data, contract term data, and / or service call parameters are checked for deviations according to the deviation check rules, and a risk rating alert is generated based on the deviation check results.

[0050] Among them, deviation verification rules refer to rules used to determine whether data deviates from the preset range or standard strategy. Deviation verification rules may include price verification rules, term verification rules and / or parameter verification rules. Accordingly, price deviation threshold verification can be performed on price configuration data based on price verification rules, or term deviation threshold verification can be performed on contract term data based on term verification rules, or parameter deviation threshold verification can be performed on service call parameters based on parameter verification rules.

[0051] Specifically, the deviation verification rules corresponding to price configuration data, contract term data, and / or service call parameters can first be read from the rule base. Based on the obtained deviation verification rules, the deviation verification of each data item is performed according to the deviation verification rules, and a risk rating reminder is generated based on the deviation verification results.

[0052] In practical applications, contract documents may contain issues such as ambiguous wording and non-standardized text. In such cases, after the intelligent agent parses the contract documents and processes the contract data, problems such as misidentification of information and incorrect character matching may occur. To address this, the contract parsing data or contract data can be converted into dialogue data and visualized through a dialogue component. This allows operations personnel to intuitively view the contract parsing data or contract data through dialogue, and to provide modification suggestions in the dialogue, thereby further calling the corresponding contract processing tools for secondary processing. In one optional implementation of this embodiment, after the intelligent agent's toolset calls the corresponding contract processing tool to process the contract parsing data and obtain the contract data, it further includes: Generate contract parsing data or dialog data of contract data, and display the dialog data through a dialog component; If dialogue content submitted through the dialogue component is detected, the contract data will be processed according to the contract processing tool corresponding to the dialogue content.

[0053] Specifically, based on the obtained contract parsing data or contract data, corresponding dialogue data can be generated and displayed to operations personnel through a dialogue component in the interactive interface. Subsequently, if operations personnel submit dialogue interaction content in the dialogue component, such as feedback, modification opinions, and / or confirmation instructions, the corresponding contract processing tool in the intelligent agent toolset will be invoked upon detecting the dialogue interaction content. The tool will process the current contract parsing data or contract data according to the dialogue interaction content and obtain the processed data, such as correcting, supplementing, or reprocessing the contract parsing data or contract data. Here, the processed data obtained from the contract data processing can be displayed in the dialogue component, and operations personnel can further submit dialogue interaction content based on the displayed processed data.

[0054] Following the previous example, after calling the risk assessment tool to perform a risk rating on the contract term data, the risk rating result can be generated as contract data, and the dialogue data of the contract data can be generated: "The contract renewal period is inconsistent with the content of the multimodal contract document. It was detected that the contract renewal period is a years, while the multimodal contract document stipulates that the contract renewal period shall not exceed b years (a>b), which may pose a performance risk." The dialogue data is displayed through the dialogue component. If the dialogue interaction content submitted by the operations personnel through the dialogue component is detected, "Please change the contract renewal period to b years", the contract term data is modified according to the contract term correction tool corresponding to the dialogue interaction content.

[0055] Furthermore, in the process of parsing or processing contract data, it is not always necessary to use contract processing tools. In this case, processing contract data according to the contract processing tools corresponding to the dialogue interaction content can be replaced by processing contract data according to the contract processing operations corresponding to the dialogue interaction content.

[0056] In the specific execution process, after processing the contract data, it can jump to the next contract processing node, for example, it can also generate a service call protocol; specifically, during the contract processing process, the corresponding contract data can be obtained from multiple contract processing nodes of the credit reporting service, such as the credit reporting product selection node, the contract price generation node, and / or the credit reporting service signing node; based on this, a service call protocol can be generated based on the contract data of at least one contract processing node; optionally, the service call protocol is generated based on the contract data of multiple contract processing nodes of the credit reporting service.

[0057] Subsequently, after generating the service invocation protocol, upon detecting a service client's invocation request for the credit reporting service, service invocation processing can be performed based on the service invocation protocol; in an optional implementation provided in this embodiment, it further includes: If a service client's call request for the credit reporting service is detected, the call request is checked based on the service call protocol, and the service call is processed after the check passes.

[0058] Specifically, if a service client's call request for the credit scoring service is detected, the call request can be intercepted and the call request can be checked based on the service call protocol. If the call check passes, the service call can be processed; otherwise, if the call check fails, no processing or an error will be returned.

[0059] It should be noted that the above-described methods for processing contract parsing data to obtain contract data can be combined in any form according to actual execution needs. Alternatively, they can be adapted, modified, or deleted and then combined in any form according to actual execution needs. For example, processing contract parsing data to obtain contract data includes: extracting form data from the contract parsing data according to the configuration form corresponding to the contract processing node and filling it into the configuration form to obtain the configuration form data of the contract processing node; calling a verification tool to perform consistency verification between the multimodal contract file and the price configuration data, contract term data, and service call parameters, and obtaining the verification result. Alternatively, contract data can be obtained by processing the contract parsing data, including: reading the deviation verification rules corresponding to the price configuration data, contract term data, and service call parameters; performing deviation verification on the price configuration data, contract term data, and service call parameters according to the deviation verification rules; generating a risk rating alert based on the deviation verification results; and displaying a confidence level prompt on the interactive page based on the confidence level of the contract parsing data output by the large language model. Furthermore, based on the arbitrary combination of implementation methods for processing contract data to obtain contract data from contract parsing data, it is also possible to make adaptive combinations by combining the aforementioned arbitrary combinations of implementation methods for parsing content of multimodal contract documents. Furthermore, the above-mentioned implementation method for processing contract parsing data to obtain contract data can be combined in any form according to the needs of the contract processing nodes in the actual execution process. Alternatively, it can be adapted, modified, or deleted according to the needs of the contract processing nodes in the actual execution process before being combined in any form. For example, processing contract parsing data to obtain contract data includes: extracting form data from the contract parsing data according to the configuration form corresponding to the first contract processing node and filling it into the configuration form to obtain the configuration form data of the first contract processing node; calling a verification tool to perform consistency verification between the multimodal contract file and the price configuration data of the second contract node to obtain the verification result, and calling a risk assessment tool to perform risk rating on the price configuration data of the second contract node and providing a risk rating reminder on the interactive page; calling a verification tool to perform consistency verification between the multimodal contract file and the service call parameters of the third contract node to obtain the verification result, and calling a risk assessment tool to perform risk rating on the service call parameters of the third contract node and providing a risk rating reminder on the interactive page.

[0060] Furthermore, it should be noted that the contract processing methods for different contract processing nodes provided above can be combined in any form according to actual execution needs. During the combination process, the corresponding implementation method can be selected and combined with the implementation methods of one or more other processing nodes according to the execution needs of different contract processing nodes. In specific execution, the multimodal contract file uploaded by the first contract processing node of the credit reporting service can be obtained first; the content of the multimodal contract file uploaded by the first contract processing node can be parsed through the large language model integrated by the intelligent agent to obtain the contract parsing data of the first contract processing node; the corresponding contract processing tool can be called in the toolset of the intelligent agent to process the contract parsing data of the first contract processing node to obtain the contract data. Subsequently, based on the contract data obtained from the first contract processing node, the multimodal contract file uploaded by the second contract processing node of the credit reporting service can also be obtained; the content of the multimodal contract file uploaded by the second contract processing node is parsed through the large language model integrated by the intelligent agent to obtain the contract parsing data of the second contract processing node; the corresponding contract processing tool is called in the intelligent agent's toolset to process the contract parsing data of the second contract processing node to obtain contract data, so as to generate the service call protocol of the credit reporting service based on the contract data of the first and second contract processing nodes; In this context, the first contract processing node can be any one of the following processing nodes: credit product selection node, contract price generation node, and credit service signing node. Correspondingly, the process of parsing the multimodal contract document and processing the contract data in the first contract processing node can be any one of the above-mentioned implementation methods for parsing the multimodal contract document and processing the contract data in the above-mentioned implementation methods. Similarly, the second contract processing node can be any one of the above-mentioned credit product selection node, contract price generation node, and credit service signing node. Correspondingly, the content parsing process of the multimodal contract document and the contract data processing process of the contract parsing data performed at the second contract processing node can adopt any one of the above-mentioned implementation methods for content parsing of multimodal contract documents and any one of the above-mentioned implementation methods for contract data processing of contract parsing data. For example, it can acquire multimodal contract files uploaded at the contract price generation node of the credit reporting service; use the large language model integrated by the intelligent agent to parse the content of the multimodal contract files uploaded at the contract price generation node to obtain the contract parsing data of the contract price generation node; and call the corresponding contract processing tool in the intelligent agent's toolset to process the contract parsing data of the contract price generation node to obtain the contract data. Subsequently, the process involves: acquiring the multimodal contract files uploaded by the credit reporting service signing node; parsing the content of the multimodal contract files uploaded by the credit reporting service signing node using a large language model integrated by the intelligent agent to obtain the contract parsing data of the credit reporting service signing node; calling the corresponding contract processing tool in the intelligent agent's toolset to process the contract parsing data of the credit reporting service signing node to obtain the contract data; and generating a service call agreement based on the contract data of the signing price generation node and the contract data of the credit reporting service signing node.

[0061] In summary, the agent-based contract processing method provided in this embodiment, during the contract processing process, after obtaining the multimodal contract file uploaded to the contract processing node of the credit reporting service, calls an agent to process the multimodal contract file. The agent responds to the call by inputting the multimodal contract file into a large language model for content parsing, obtaining the contract parsing data from the contract processing node, and then calling the corresponding contract processing tool in the toolset to process the contract parsing data and obtain the contract data. Subsequently, the agent returns the obtained contract data to the server. Based on this, the agent obtains the contract data and generates a service call protocol for the credit reporting service, thereby improving the efficiency and accuracy of the contract processing and signing processes.

[0062] The following example uses the application of an agent-based contract processing method provided in this embodiment in a first credit reporting service scenario, combined with... Figure 3 The agent-based contract processing method provided in this embodiment will be further explained below. Figure 3 The agent-based contract processing method applied to the first credit reporting service scenario includes the following steps.

[0063] Step S302: Obtain the multimodal contract file uploaded at the first contract processing node of the credit reporting service.

[0064] Step S304: Perform semantic recognition on the multimodal contract documents and determine the contract signing object based on the semantic recognition results.

[0065] Step S306: Identify the credit reporting products in the multimodal contract documents, and extract price data according to the price template corresponding to the credit reporting products to obtain price configuration data.

[0066] Step S308: Parse the service call parameters of the multimodal contract file to obtain the service call parameters of the service client.

[0067] Optionally, service call parameters may include service response time, data update parameters, and / or service priority.

[0068] Step S310: Extract form data from the contract parsing data according to the configuration form corresponding to the first contract processing node and fill it into the configuration form to obtain the configuration form data of the first contract processing node.

[0069] Step S312: Call the verification tool to perform consistency verification between the multimodal contract file, price configuration data, and service call parameters, obtain the verification result, and use the verification result as contract data.

[0070] Step S314: Generate dialogue data for the contract data and display the dialogue data through the dialogue component.

[0071] Step S316: Process the contract data according to the contract processing tool corresponding to the dialogue interaction content submitted through the dialogue component to obtain the processed contract data, and generate the service call protocol for the credit reporting service based on the processed contract data.

[0072] After step S316 is executed, the service call agreement can be approved after it is generated to verify the consistency between the service call agreement and the multimodal contract document.

[0073] Step S318: Based on the service call protocol, perform call detection on the call request of the service client for the credit scoring service, and process the service call after the detection is passed.

[0074] It should be noted that any one or more steps in steps S302 to S318 can be combined with any one or more steps in steps S202 to S206 to form a new implementation method according to the needs of implementation and deployment. In addition, any one or more technical features in steps S302 to S318 can be selected and combined with any one or more technical features provided in steps S202 to S206 to form a new implementation method according to the actual deployment needs. Alternatively, any one or more technical features in steps S302 to S318 can be replaced with any one or more technical features provided in steps S202 to S206 to form a new implementation method according to the actual deployment needs. These will not be elaborated on here.

[0075] The following example uses the application of an agent-based contract processing method provided in this embodiment in a second credit reporting service scenario, combined with... Figure 4 The agent-based contract processing method provided in this embodiment will be further explained below. Figure 4 The agent-based contract processing method applied to the second credit reporting service scenario includes the following steps.

[0076] Step S402: Obtain the multimodal contract file uploaded at the first contract processing node of the credit reporting service.

[0077] Step S404: The multimodal contract file of the first contract processing node is parsed using the large language model integrated by the intelligent agent to obtain the first contract parsing data of the first contract processing node.

[0078] Step S406: In the toolset of the intelligent agent, call the corresponding contract processing tool to process the first contract parsing data of the first contract processing node to obtain the first contract data.

[0079] Step S408: Obtain the multimodal contract file uploaded at the second contract processing node of the credit reporting service.

[0080] Step S410: The multimodal contract file of the second contract processing node is parsed using the large language model integrated by the intelligent agent to obtain the second contract parsing data of the second contract processing node.

[0081] Step S412: In the toolset of the intelligent agent, call the corresponding contract processing tool to process the contract data of the second contract parsing data of the second contract processing node to obtain the second contract data.

[0082] Step S414: Generate a service call protocol based on the first contract data of the first contract processing node and the second contract data of the second contract processing node of the credit reporting service.

[0083] Step S416: Based on the service call protocol, perform call detection on the call request of the service client for the credit scoring service, and process the service call after the detection is passed.

[0084] It should be noted that any one or more steps in steps S402 to S416 can be combined with any one or more steps in steps S202 to S206 to form a new implementation method according to the needs of implementation and deployment. In addition, any one or more technical features in steps S402 to S416 can be selected and combined with any one or more technical features provided in steps S202 to S206 to form a new implementation method according to the actual deployment needs. Alternatively, any one or more technical features in steps S402 to S416 can be replaced with any one or more technical features provided in steps S202 to S206 to form a new implementation method according to the actual deployment needs. These will not be elaborated on here.

[0085] This specification provides an embodiment of a contract processing device based on intelligent agents, as follows: In the above embodiments, a contract processing method based on intelligent agents is provided, and correspondingly, a contract processing device based on intelligent agents is also provided, which will be described below with reference to the accompanying drawings.

[0086] Reference Figure 5 This illustration shows a schematic diagram of an embodiment of a contract processing device based on an intelligent agent provided in this embodiment.

[0087] Since the apparatus embodiments correspond to the method embodiments, the descriptions are relatively simple. For relevant parts, please refer to the corresponding descriptions of the method embodiments provided above. The apparatus embodiments described below are merely illustrative.

[0088] This embodiment provides a contract processing device based on an intelligent agent, the device comprising: The file acquisition module 502 is configured to acquire multimodal contract files uploaded at the contract processing node of the credit reporting service; The content parsing module 504 is configured to parse the multimodal contract file using a large language model integrated by the intelligent agent, and obtain the contract parsing data of the contract processing node. The protocol generation module 506 is configured to call the corresponding contract processing tool in the toolkit of the intelligent agent to process the contract parsing data to obtain contract data, so as to generate the service call protocol of the credit reporting service.

[0089] For ease of description, the above devices are described by dividing them into various modules or units based on their functions. Of course, when implementing one or more of these specifications, the functions of each module or unit can be implemented in the same or different software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0090] This specification provides an example of a contract processing device based on intelligent agents, as follows: Corresponding to the agent-based contract processing method described above, and based on the same technical concept, one or more embodiments of this specification also provide an agent-based contract processing device, which is used to execute the agent-based contract processing method provided above. Figure 6 This is a schematic diagram of the structure of a contract processing device based on an intelligent agent, provided for one or more embodiments of this specification.

[0091] This embodiment provides a contract processing device based on intelligent agents, comprising: like Figure 6 As shown, device 600 mainly consists of a communication interface 602, a user interface 604, a processor 606, and a data storage 608. These components are interconnected and communicate with each other via a system bus, network, or other connection mechanism 610. The communication interface 602 enables device 600 to communicate with other devices, access networks, and transmission networks via analog or digital modulation. For example, the communication interface 602 may include a chipset and antenna for wireless communication with a radio access network or access point. Furthermore, the communication interface 602 can be a wired interface such as Ethernet, Token Ring, or a USB port, or a wireless interface such as Wi-Fi, Bluetooth, Global Positioning System (GPS), or a wide-area wireless interface (e.g., WiMAX or LTE). Of course, the communication interface 602 can also support other forms of physical layer interfaces and standard or proprietary communication protocols. The communication interface 602 may also include multiple physical communication interfaces, such as Wi-Fi, Bluetooth, and wide-area wireless interfaces. The user interface 604 includes receiving user input and providing output to the user. Therefore, the user interface 604 may include input components such as a keypad, keyboard, touch-sensitive or presence-sensitive panel, computer mouse, trackball, joystick, microphone, still camera, and video camera, and output components such as a display screen (which may be combined with a touch-sensitive panel), CRT, LCD, LED, display using DLP technology, printer, and other similar devices known or developed in the future. The user interface 604 may also generate auditory output via speakers, speaker jacks, audio output ports, audio output devices, headphones, and other similar devices known or developed in the future. In some embodiments, the user interface 604 may include software, circuitry, or other forms of logic capable of transmitting and receiving data from external user input / output devices. Additionally or alternatively, the device 600 may support remote access from other devices via communication interface 602 or another physical interface (not shown). The user interface 604 may be configured to receive input, the position and movement of which may be indicated by indicators or cursors described herein. The user interface 604 may also be configured as a display device for rendering or displaying text fragments.

[0092] Processor 606 may include one or more general-purpose processors and / or special-purpose processors. Data storage 608 may include one or more volatile and / or non-volatile storage components and may be integrated wholly or partially with processor 606. Data storage 608 may include removable and non-removable components.

[0093] Processor 606 is capable of executing program instructions 618 (e.g., compiled or uncompiled program logic and / or machine code) stored in data storage 608 to perform the various functions described herein. Data storage 608 may comprise a non-transitory computer-readable medium on which program instructions are stored, which, when executed by device 600, enable device 600 to perform any methods, processes, or functions disclosed in this specification and / or the accompanying drawings. Execution of program instructions 618 by processor 606 may result in processor 606 using data 612. For example, program instructions 618 may include an operating system 622 (e.g., an operating system kernel, device drivers, and / or other modules) installed on device 600 and one or more application programs 620 (e.g., a browser, social application, or game application). Similarly, data 612 may include operating system data 616 and application data 614. Operating system data 616 is primarily accessible to operating system 622, while application data 614 is primarily accessible to one or more application programs 620. Application data 614 may reside in a file system visible or hidden from the user of device 600. Application 620 can communicate with operating system 622 through one or more application programming interfaces (APIs). These APIs facilitate application 620 in reading and / or writing application data 614, transmitting or receiving information via communication interface 602, and receiving or displaying information on user interface 604. In some terms, application 620 may be simply referred to as "app". Furthermore, application 620 can be downloaded to device 600 through one or more online app stores or app markets. However, applications can also be installed on device 600 in other ways, such as through a web browser or a physical interface on device 600 (e.g., a USB port).

[0094] In one specific embodiment, the agent-based contract processing device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for use in a server, and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following: Retrieve multimodal contract files uploaded at the contract processing node of the credit reporting service; The multimodal contract file is parsed using a large language model integrated by an intelligent agent to obtain the contract parsing data of the contract processing node; The intelligent agent's toolset invokes the corresponding contract processing tool to process the contract parsing data and obtain contract data, thereby generating the service invocation protocol for the credit reporting service.

[0095] This specification provides an embodiment of a computer-readable storage medium as follows: Corresponding to the agent-based contract processing method described above, and based on the same technical concept, one or more embodiments of this specification also provide a computer-readable storage medium.

[0096] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions, which, when executed, implement the following process: Retrieve multimodal contract files uploaded at the contract processing node of the credit reporting service; The multimodal contract file is parsed using a large language model integrated by an intelligent agent to obtain the contract parsing data of the contract processing node; The intelligent agent's toolset invokes the corresponding contract processing tool to process the contract parsing data and obtain contract data, thereby generating the service invocation protocol for the credit reporting service.

[0097] It should be noted that the embodiments of a computer-readable storage medium described in this specification and the embodiments of a contract processing method based on an intelligent agent described in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0098] This specification provides an example of a computer program product as follows: Corresponding to the agent-based contract processing method described above, and based on the same technical concept, one or more embodiments of this specification also provide a computer program product.

[0099] A computer program product includes a computer program / instructions that, when executed by a processor, perform the following steps: Retrieve multimodal contract files uploaded at the contract processing node of the credit reporting service; The multimodal contract file is parsed using a large language model integrated by an intelligent agent to obtain the contract parsing data of the contract processing node; The intelligent agent's toolset invokes the corresponding contract processing tool to process the contract parsing data and obtain contract data, thereby generating the service invocation protocol for the credit reporting service.

[0100] It should be noted that the embodiments of a computer program product described in this specification and the embodiments of a contract processing method based on an intelligent agent described in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can be referred to the implementation of the corresponding method described above, and the repeated parts will not be described again.

[0101] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. Each embodiment focuses on describing the differences from other embodiments. For example, the device embodiments, equipment embodiments, computer-readable storage medium embodiments, and computer program product embodiments are all similar to the method embodiments, so the descriptions are relatively simple. For reading the relevant content of the device embodiments, equipment embodiments, computer-readable storage medium embodiments, and computer program product embodiments, please refer to the description of the method embodiments.

[0102] While one or more embodiments of this specification provide method steps as described in the embodiments or flowcharts, it is understood that the order of steps listed in the embodiments or flowcharts is merely one possible execution order among many steps, and does not represent the only execution order. Therefore, when the claims involve method steps, modifications to the order of such steps, or parallel execution between steps, are also within the scope of protection of the claims. This specification uses specific terms to describe embodiments of this specification. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this specification do not necessarily refer to the same embodiment. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0103] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0104] In the 1930s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many improvements to the methodology today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that an improvement to the methodology cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0105] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0106] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0107] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0108] Those skilled in the art will understand that one or more embodiments of this specification can be provided as a method, system, or computer program product. Therefore, one or more embodiments of this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. 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 illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0110] 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.

[0111] 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.

[0112] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0113] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0114] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0115] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising at least one…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0116] One or more embodiments of this specification can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. One or more embodiments of this specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0117] The above description is merely an embodiment of this document and is not intended to limit the scope of this document. Various modifications and variations can be made to this document by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this document should be included within the scope of the claims of this document.

Claims

1. A contract processing method based on intelligent agents, comprising: Retrieve multimodal contract files uploaded at the contract processing node of the credit reporting service; The multimodal contract file is parsed using a large language model integrated by an intelligent agent to obtain the contract parsing data of the contract processing node; The intelligent agent's toolset invokes the corresponding contract processing tool to process the contract parsing data and obtain contract data, thereby generating the service invocation protocol for the credit reporting service.

2. The agent-based contract processing method according to claim 1, wherein parsing the content of the multimodal contract document includes: The multimodal contract documents are subjected to semantic recognition, and the contract signing object is determined based on the semantic recognition results; Based on the semantic recognition results, the signing scheme of the credit reporting product is extracted, and the credit reporting product data of the credit reporting product included in the signing scheme is extracted.

3. The contract processing method based on intelligent agents according to claim 2, wherein the contract data processing is implemented in the following manner: According to the configuration form corresponding to the contract processing node, extract form data from the contract parsing data and populate it into the configuration form to obtain the configuration form data of the contract processing node; The confidence level of the contract parsing data output by the large language model is displayed on the interactive page.

4. The contract processing method based on an intelligent agent according to claim 1, further comprising, after the step of calling the corresponding contract processing tool in the intelligent agent's toolset to process the contract parsing data and obtain contract data, the method includes: Generate the contract parsing data or the dialogue data of the contract data, and display the dialogue data through a dialogue component; If dialogue content submitted through the dialogue component is detected, the contract data is processed according to the contract processing tool corresponding to the dialogue content.

5. The contract processing method based on intelligent agents according to claim 1, wherein the service invocation protocol is generated based on contract data from multiple contract processing nodes of the credit reporting service; If a service client's call request for the credit reporting service is detected, the call request is checked based on the service call protocol, and the service call is processed after the check passes.

6. The agent-based contract processing method according to claim 1, wherein parsing the content of the multimodal contract file includes: Identify the credit reporting products and / or contracting schemes in the multimodal contract documents, and extract price configuration data according to the price templates corresponding to the credit reporting products and / or contracting schemes.

7. The agent-based contract processing method according to claim 6, wherein the content parsing of the multimodal contract document further includes: Contract term data is obtained by extracting the contract term from the semantic recognition results of the multimodal contract document; And / or, The service call parameters of the service client are obtained by parsing the service call parameters of the multimodal contract file; the service call parameters include service response time, data update parameters and / or service priority.

8. The agent-based contract processing method according to claim 6 or 7, wherein the step of calling the corresponding contract processing tool to process the contract parsing data to obtain contract data includes: The verification tool is invoked to verify the consistency between the multimodal contract file and the price configuration data, the contract term data, and / or the service call parameters, and the verification result is obtained. The risk assessment tool is invoked to perform a risk rating on the price configuration data, the contract term data, and / or the service call parameters, and a risk rating reminder is displayed on the interactive page.

9. The agent-based contract processing method according to claim 8, wherein the risk rating of the price configuration data, the contract term data, and / or the service invocation parameters includes: Read the deviation verification rules corresponding to the price configuration data, the contract term data, and / or the service call parameters; According to the deviation verification rules, the price configuration data, the contract term data, and / or the service call parameters are checked for deviations, and a risk rating alert is generated based on the deviation verification results.

10. The agent-based contract processing method according to claim 1, wherein the contract processing tool is determined in the following manner: The contract processing tool corresponding to the contract processing node is determined in the toolset based on the contract processing node, or the file type of the multimodal contract file and / or the contract parsing data corresponding to the contract processing tool is determined in the toolset.

11. The agent-based contract processing method according to claim 1, wherein the multimodal contract file comprises: The signed contract document obtained by the credit reporting platform and its service clients through at least one signed agreement or at least one credit reporting contract; The signing methods for the aforementioned signing scheme or credit investigation contract include offline signing and / or online signing.

12. A contract processing device based on intelligent agents, comprising: The file acquisition module is configured to acquire multimodal contract files uploaded at the contract processing node of the credit reporting service; The content parsing module is configured to parse the multimodal contract file using a large language model integrated by the intelligent agent, and obtain the contract parsing data of the contract processing node. The protocol generation module is configured to call the corresponding contract processing tool in the toolkit of the intelligent agent to process the contract parsing data to obtain contract data, so as to generate the service call protocol of the credit reporting service.

13. A contract processing device based on intelligent agents, comprising: processor; And, a memory configured to store computer-executable instructions, which, when executed, cause the processor to: Retrieve multimodal contract files uploaded at the contract processing node of the credit reporting service; The multimodal contract file is parsed using a large language model integrated by an intelligent agent to obtain the contract parsing data of the contract processing node; The intelligent agent's toolset invokes the corresponding contract processing tool to process the contract parsing data and obtain contract data, thereby generating the service invocation protocol for the credit reporting service.

14. A computer-readable storage medium for storing computer-executable instructions that, when executed, implement the steps of the method of claim 1.

Images