An auxiliary case-handling system, method, and apparatus

Abstract

The embodiment of the invention discloses an auxiliary case handling system, method and device. A task planning module determines a next planning task according to current state information of a target case, the current state information of the target case comprises case information and / or an intermediate execution result, and the next planning task comprises generation of a subtask and / or generation of an initial execution strategy. At least one task execution module executes the subtasks generated by the task planning module to obtain the intermediate execution result and sends the intermediate execution result to the task planning module, and a strategy output module obtains strategy display information according to the initial execution strategy generated by the task planning module and outputs the strategy display information. Therefore, automatic auxiliary processing of the case can be realized, the processing efficiency of the litigation case is improved, the case processing time is shortened, and the labor cost is reduced.

Description

(19) State Intellectual Property Office (12) Invention Patent Application (10) Application Publication Number (43) Application Publication Date (21) Application Number 202511604225.2 (22) Application Date 2025.11.04 (71) Applicant Alibaba (China) Co., Ltd. Address 310052, Room 508, 5th Floor, Building 4, No. 699, Wangshang Road, Changhe Street, Binjiang District, Hangzhou City, Zhejiang Province (72) Inventor Jin Shiyang (74) Patent Agency Beijing Ruipai Intellectual Property Agency Co., Ltd. 11597 Patent Attorney Liu Feng (51) Int.Cl. G06Q 50 / 18 (2012.01) G06Q 10 / 10 (2023.01) (54) Invention Title A Case-Handling Assistance System, Method and Apparatus (57) Abstract The embodiments of this invention disclose a case-handling assistance system, method and apparatus. The task planning module determines the next planned task based on the current status information of the target case, including case information and / or intermediate execution results. The next planned task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module executes the sub-tasks generated by the task planning module to obtain the intermediate execution results and sends them back to the task planning module. The strategy output module obtains and outputs strategy display information based on the initial execution strategy generated by the task planning module. This enables automated auxiliary processing of cases, improves the efficiency of litigation case handling, shortens case processing time, and reduces labor costs. Claims 3 pages, Description 11 pages, Drawings 7 pages, CN 121563719 A 2026.02.24 CN 1 21 56 37 19 A 1. An auxiliary case-handling system, characterized in that the system comprises: a task planning module, used to determine the next planning task based on the current status information of the target case, wherein the current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating an initial execution strategy; at least one task execution module, used to execute the sub-tasks generated by the task planning module to obtain the intermediate execution results, and send the intermediate execution results to the task planning module; a strategy output module, used to obtain strategy display information based on the initial execution strategy generated by the task planning module and output it. 2. The system according to claim 1, characterized in that the task planning module, at least one task execution module, and strategy output module are intelligent agents. 3. The system according to claim 2, characterized in that the system further comprises: a memory module for storing long-term memory and short-term memory; wherein the task planning module, at least one task execution module, and strategy output module are used to retrieve the long-term memory and short-term memory from the memory module.4. The system according to claim 1, characterized in that the system further comprises: a human-computer interaction module, used to receive case information input by a user, the case information including questions input by the user and / or litigation documents, the litigation documents including documents submitted by the parties and / or documents prepared by the court; and / or, receiving and displaying strategy display information sent by the strategy output module. 5. The system according to claim 4, characterized in that the subtask includes an instruction and a target task execution module. 6. The system according to claim 5, characterized in that the system further comprises at least one of the following: a case management module, used to manage various cases; a knowledge storage module, used to store normative documents, the normative documents including one or more of laws and regulations, platform rules, and regulations; a data storage module, used to store data; and a tool center, including at least one access tool for retrieving data from the data storage module. 7. The system according to claim 6, wherein the task execution module comprises at least one of the following: a data query module, configured to invoke a corresponding access tool to retrieve relevant data from the data storage module according to the instruction; a knowledge query module, configured to retrieve corresponding normative documents from the knowledge storage module according to the instruction; an evidence collection module, configured to retrieve our evidence according to the instruction; a delegation module, configured to retrieve a recommended execution strategy according to the instruction, the recommended execution strategy including settlement, mediation, and court hearing; a registration module, configured to enter litigation documents into the case management module according to the instruction; and a risk diagnosis module, configured to retrieve the general risk assessment result of the target case according to the instruction. 8. The system according to claim 7, characterized in that: the task planning module is used to determine the processing type of the target case according to the case information, the processing type including strategy generation and information registration; in response to the processing type of the target case being strategy generation, a knowledge query sub-task is generated and sent to the knowledge query module; the knowledge query module is used to execute the knowledge query sub-task to obtain normative documents; the task planning module is used to determine the focus of the target case according to the normative documents, and generate an investigation sub-task according to the focus; the investigation module is used to execute the investigation sub-task to obtain our evidence; the task planning module is used to generate a risk diagnosis sub-task; the risk diagnosis module is used to execute the risk diagnosis sub-task to obtain a general risk assessment result of the target case; the task planning module is used to determine the overall risk assessment result of the target case according to one or more of the case information, focus, our evidence, and general risk assessment result, and generate a strategy generation sub-task according to the overall risk assessment result;The delegation module is used to execute the strategy generation sub-task to obtain the recommended execution strategy for the target case; the task planning module is used to generate the initial execution strategy according to the recommended execution strategy. 9. The system according to claim 8, wherein the task planning module is further used to generate an information registration sub-task and send it to the registration module in response to the processing type of the target case being information registration; the registration module is used to execute the information registration sub-task to enter litigation documents in the case management module. 10. The system according to claim 1, wherein the strategy output module is used to convert the initial execution strategy into structured data, extract key information from the structured data using a large language model, generate intermediate content according to the key information, and adjust the intermediate content to obtain the strategy display information. 11. An auxiliary case handling method, wherein the method includes: determining the next planning task according to the current status information of the target case, wherein the current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating an initial execution strategy; executing the sub-tasks to obtain the intermediate execution results; obtaining strategy display information according to the initial execution strategy and outputting it. 12. The method according to claim 11, characterized in that the method further comprises: receiving case information input by a user, the case information including questions input by the user and / or litigation documents, the litigation documents including documents submitted by the parties and / or documents prepared by the court. 13. The method according to claim 11, characterized in that the subtask includes an instruction and a target task execution module. 14. The method according to claim 13, characterized in that executing the subtask to obtain the intermediate execution result includes at least one of the following: calling a corresponding access tool to obtain relevant data from a data storage module according to the instruction; obtaining corresponding normative documents from a knowledge storage module according to the instruction; obtaining our evidence according to the instruction; obtaining a recommended execution strategy according to the instruction, the recommended execution strategy including settlement, mediation, and court hearing; entering litigation documents into a case management module according to the instruction; and obtaining a general risk assessment result for the target case according to the instruction. 15. The method according to claim 14, characterized in that, determining the next planning task based on the current status information of the target case includes: determining the processing type of the target case based on the case information, the processing type including strategy generation and information registration; in response to the processing type of the target case being strategy generation, generating a knowledge query subtask to obtain normative documents; and determining the focus of the target case based on the normative documents.Based on the stated concerns, generate an investigation sub-task to obtain our evidence; generate a risk diagnosis sub-task to obtain a general risk assessment result for the target case; determine the overall risk assessment result for the target case based on one or more of the case information, concerns, our evidence, and general risk assessment result; generate a strategy generation sub-task based on the overall risk assessment result to obtain a recommended execution strategy for the target case; generate the initial execution strategy based on the recommended execution strategy. 16. The method according to claim 15, wherein determining the next planning task based on the current status information of the target case further includes: in response to the processing type of the target case being information registration, generating an information registration sub-task to enter litigation documents in the case management module. 17. The method according to claim 11, wherein obtaining and outputting strategy display information based on the initial execution strategy includes: converting the initial execution strategy into structured data; extracting key information from the structured data using a large language model; generating intermediate content based on the key information; and adjusting the intermediate content to obtain the strategy display information. 18. An auxiliary case-handling device, characterized in that the device comprises: a planning unit, configured to determine a next planning task based on the current status information of a target case, wherein the current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating an initial execution strategy; an execution unit, configured to execute the sub-tasks to obtain the intermediate execution results; and a display unit, configured to obtain and output strategy display information based on the initial execution strategy. 19. An electronic device, comprising a memory and a processor, characterized in that the memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method as described in any one of claims 11-17. 20. A computer-readable storage medium storing computer program instructions thereon, characterized in that the computer program instructions, when executed by a processor, implement the method as described in any one of claims 11-17. 21. A computer program product, comprising a computer program, characterized in that when the computer program is run on a computer, the computer performs the method as described in any one of claims 11-17. Claims 3 / 3 Page 4 CN 121563719 A A Case-Helping System, Method and Apparatus Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a case-handling system, method and apparatus. Background Art

[0002] Efficient and standardized handling of legal cases is crucial for various entities to control legal risks and protect their legitimate rights and interests.This includes not only administrative tasks such as case registration, document drafting, and material transfer, but also extends to the core litigation strategy determination stage, such as judging key decisions like settlement, mediation, court hearings, or appeals based on the case facts.

[0003] Currently, case handling relies heavily on manual processing. Professional human resources are not only consumed by a large number of repetitive tasks, but also require strategic analysis for each case, resulting in low overall case handling efficiency, long cycles, and high costs. Summary of the Invention

[0004] In view of this, the purpose of the embodiments of the present invention is to provide an auxiliary case handling system, method, and apparatus that can improve the efficiency of litigation case handling and reduce labor costs.

[0005] In a first aspect, embodiments of the present invention provide an auxiliary case-handling system, the system comprising: a task planning module, configured to determine a next planning task based on the current status information of a target case, wherein the current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating an initial execution strategy; at least one task execution module, configured to execute the sub-tasks generated by the task planning module to obtain the intermediate execution results, and send the intermediate execution results to the task planning module; and a strategy output module, configured to obtain and output strategy display information based on the initial execution strategy generated by the task planning module.

[0006] In a second aspect, embodiments of the present invention provide an auxiliary case-handling method, the method comprising: determining a next planning task based on the current status information of a target case, wherein the current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating an initial execution strategy; executing the sub-tasks to obtain the intermediate execution results; and obtaining and outputting strategy display information based on the initial execution strategy.

[0007] In a third aspect, embodiments of the present invention provide an auxiliary case-handling device, the device comprising: a planning unit, configured to determine a next planning task based on the current status information of a target case, the current status information of the target case including case information and / or intermediate execution results, the next planning task including generating sub-tasks and / or generating an initial execution strategy; an execution unit, configured to execute the sub-tasks to obtain the intermediate execution results; and a display unit, configured to obtain and output strategy display information based on the initial execution strategy.

[0008] In a fourth aspect, embodiments of the present invention provide an electronic device, including a memory and a processor, the memory being configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method as described in the first aspect. Specification 1 / 11 pages 5 CN 121563719 A

[0009] In a fifth aspect, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon.The computer program instructions, when executed by a processor, implement the method described in the first aspect.

[0010] In a sixth aspect, embodiments of the present invention provide a computer program product comprising a computer program that, when run on a computer, executes the method described in the first aspect.

[0011] The technical solution of the embodiments of the present invention determines the next planning task based on the current status information of the target case by a task planning module. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module executes the sub-tasks generated by the task planning module to obtain the intermediate execution results and sends the intermediate execution results to the task planning module. The strategy output module obtains strategy display information based on the initial execution strategy generated by the task planning module and outputs it. Thus, automated auxiliary processing of cases can be achieved, improving the efficiency of litigation case processing, shortening case processing time, and reducing labor costs.

[0012] The above and other objects, features and advantages of the present invention will become clearer from the following description of embodiments of the present invention with reference to the accompanying drawings, in which: FIG1 is a schematic diagram of an auxiliary case-handling system according to an embodiment of the present invention; FIG2 is a schematic diagram of an auxiliary case-handling system according to another embodiment of the present invention; FIG3 is a flowchart of an auxiliary case-handling method according to an embodiment of the present invention; FIG4 is a flowchart of determining the next step plan according to an embodiment of the present invention; FIG5 is a flowchart of obtaining strategy display information according to an embodiment of the present invention; FIG6 is a schematic diagram of an auxiliary case-handling device according to an embodiment of the present invention; FIG7 is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Description

[0013] The present application is described below based on embodiments, but the present application is not limited to these embodiments. In the following detailed description of the present application, some specific details are described in detail. The present application can be fully understood by those skilled in the art without these details. In order to avoid obscuring the substance of the present application, well-known methods, processes, flows, elements and circuits are not described in detail.

[0014] In addition, those skilled in the art should understand that the accompanying drawings provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0015] Unless the context explicitly requires it, the terms "comprising," "including," and similar terms used throughout the application should be interpreted as encompassing rather than exclusive or exhaustive; that is, meaning "including but not limited to."

[0016] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0017] The solutions described in this specification and embodiments, if involving the processing of personal information, will all be processed under the premise of having a legal basis (such as obtaining the consent of the personal information subject, or being necessary for the performance of a contract, etc.), and will only be processed within the scope of regulations or agreements. If a user refuses to process personal information other than the information necessary for basic functions, it will not affect the user's use of basic functions.

[0018] In view of the problems of low efficiency, long cycle and high cost of manual case handling in the prior art, the embodiments of the present invention propose an auxiliary case handling system, method and device. The auxiliary case handling system, method and device of the embodiments of the present invention can be applied to various litigation participants such as enterprises, law firms, courts, procuratorates and individual consultants. Its core functions include, but are not limited to: automatically generating execution strategy suggestions (such as settlement, mediation or court hearing) for the plaintiff or defendant based on the case, and providing support for electronic registration and management of litigation documents generated in the entire process of the case for local and cloud deployment.

[0019] Figure 1 is a schematic diagram of an auxiliary case handling system of an embodiment of the present invention. As shown in Figure 1, the auxiliary case-handling system of this embodiment includes a task planning module 1, a task execution module 2, a strategy output module 3, a memory module 4, a human-computer interaction module 5, and a data support module 6.

[0020] The human-computer interaction module 5 is used for human-computer interaction, including receiving case information output by the user and sending it to the task planning module 1; and receiving strategy display information output by the strategy output module 3 and displaying the strategy display information. The human-computer interaction module 5 is located on the user terminal side and can be implemented through applications, browser web pages, mini-programs, etc. The user terminal can be a mobile phone, laptop, desktop computer, tablet computer, etc.

[0021] The case information includes questions input by the user and / or litigation documents, and the litigation documents include documents submitted by the parties and / or documents made by the court.

[0022] For example, when a user needs to handle a certain litigation case, they input relevant questions through the human-computer interaction interface, such as "Please give the handling strategy for case number xx" or "Please give the handling strategy for case name xx", etc. Simultaneously, litigation documents can be added to the human-computer interaction interface, such as the complaint, opposing party's evidence, and judgment results (e.g., PDF, Word, etc.); or, text content can be added, that is, the text content from the complaint, opposing party's evidence, and judgment results can be copied into the user's input question.

[0023] The task planning module 1 is used to determine the next planning task based on the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy.

[0024] Specifically, the task planning module 1 analyzes and reflects on the current state information. If the current state information is insufficient to support the task planning module 1 in making an initial execution strategy, it continues to generate sub-tasks to obtain relevant information; if the current state information is sufficient to support the task planning module 1 in making an initial execution strategy, it generates an initial execution strategy and sends the initial execution strategy to the strategy output module 3.

[0025] The task execution module 2 is used to execute the sub-tasks generated by the task planning module to obtain the intermediate execution results and send the intermediate execution results to the task planning module 1. The sub-task includes an instruction and a target task execution module, wherein the target task execution module is used to indicate which task execution module will execute the sub-task, and the instruction is the detailed content of the sub-task.

[0026] In some embodiments, the task execution module 2 completes the corresponding sub-task through the data support module 6. The data support module 6 includes a tool center 61, a data storage module 62, a knowledge storage module 63, and a case management module 64.

[0027] The tool center 61 includes at least one access tool for obtaining data from the data storage module 62.

[0028] The data storage module 62 is used to store data. For example, in the e-commerce field, the data storage module can be an external database that stores merchant information, seller information, store information, etc.

[0029] The knowledge storage module 63 is used to store normative documents, which include one or more of laws and regulations, platform rules, and regulations. Specification 3 / 11 pages 7 CN 121563719 A

[0030] The case management module 64 is used to manage various cases, for example, to manage the status, documents, time limits, etc. of each case.

[0031] In some embodiments, the task execution module 2 includes one or more of the following: data query module 21, knowledge query module 22, investigation module 23, entrustment module 24, registration module 25, and risk diagnosis module 26.

[0032] Wherein, the data query module 21 is used to call the corresponding access tool to obtain relevant data from the data storage module according to the instruction. Specifically, the data query module 21 is connected to the tool center 61. After receiving the instruction sent by the task planning module 1, the data query module 21 performs tool planning according to the instruction, determines the tool to be called, and then calls the corresponding tool from the tool center 61 to obtain data from the data storage module 62.

[0033] The knowledge query module 22 is used to obtain the corresponding normative document from the knowledge storage module according to the instruction. The knowledge query module 22 is connected to the knowledge storage module 63. The knowledge query module 22 performs knowledge retrieval in the knowledge storage module 63 according to the received instruction to obtain intermediate execution results.

[0034] The evidence collection module 23 is used to obtain our evidence according to the instruction. Our evidence is pre-stored in a designated location.In the evidence system, the evidence investigation module 23 obtains our evidence by accessing the evidence system.

[0035] The entrustment module 24 is used to obtain recommended execution strategies according to the instructions. The recommended execution strategies include settlement, mediation, and court hearing.

[0036] The registration module 25 is used to enter litigation documents into the case management module 64 according to the instructions. Specifically, the registration module 25 has the right to access the case management module 64. When a sub-task is received, the relevant litigation documents are automatically registered in the case management module.

[0037] The risk diagnosis module 26 is used to obtain the general risk assessment results of the target case according to the instructions. Among them, general risk refers to risks that occur frequently in related fields. For example, in the e-commerce field, general risks include criminal risks, whether there are flaws in the customer service process, whether the platform leaks relevant information in a timely manner, whether the platform takes reasonable measures (including but not limited to: product removal or deletion, freezing of deposit, and assistance with refund services), whether the platform has no intentional or gross negligence, and whether the buyer provides evidence of authenticity. Among them, the judgment of different general risks can be implemented by a risk diagnosis module 26, or by different risk diagnosis modules 26.

[0038] The strategy output module 3 is used to obtain strategy display information and output it according to the initial execution strategy generated by the task planning module.

[0039] Specifically, the strategy output module 3 is used to convert the initial execution strategy into structured data, extract key information from the structured data using a large language model, generate intermediate content according to the key information, and adjust the intermediate content to obtain the strategy display information. The strategy output module 3 outputs the obtained strategy display information to the human-computer interaction module 5 for display.

[0040] In this embodiment of the invention, the task planning module determines the next planning task according to the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module executes the sub-tasks generated by the task planning module to obtain the intermediate execution results and sends the intermediate execution results to the task planning module. The strategy output module obtains strategy display information and outputs it according to the initial execution strategy generated by the task planning module. Therefore, automated auxiliary processing of cases can be achieved, improving the efficiency of litigation case processing, shortening case processing time, and reducing labor costs.

[0041] In some embodiments, the above-mentioned task planning module 1, task execution module 2, and strategy output module 3 can be implemented by an intelligent agent. An agent is a proxy that can perceive the environment and take actions to achieve a specific goal. An intelligent agent can be software, hardware, or a system, possessing autonomy, adaptability, and interactive capabilities. The intelligent agent perceives the environment... (The rest of the text appears to be a page from a document and doesn't need a direct translation.)The changes in 121563719 A are judged and decided based on the knowledge and algorithms learned by itself, and then the actions are executed to achieve the predetermined goal. Specifically, the task planning module 1, task execution module 2, and strategy output module 3 can all be implemented by a large model-driven intelligent agent (LLM Agent), which has functions such as understanding natural language, calling tools, and planning tasks. The large model-driven intelligent agent has functions such as perception, reasoning and planning, tool use, memory, reflection and learning. Perception refers to receiving input information such as instructions, environmental information, and historical dialogues. Reasoning and planning refers to analyzing tasks, breaking down steps, and generating action plans. Tool use refers to calling search engines, APIs (Application Programming Interfaces), databases, etc. Memory refers to storing short-term dialogue memories or long-term knowledge to support contextual learning. Reflection and learning refers to summarizing failed attempts and optimizing subsequent behaviors.

[0042] As a specific example, Figure 2 is a schematic diagram of an auxiliary case-handling system of another embodiment of the present invention. Figure 2 shows the system architecture and data processing process when the task planning module 1, task execution module 2, and strategy output module 3 are implemented by an intelligent agent.

[0043] As shown in Figure 2, after the task planning module 1 obtains the case information, it retrieves short-term and long-term memories from the memory module 4 and constructs a context, including integrating information such as user questions, short-term memories, long-term memories, and litigation documents. Based on the constructed context, the user's intent is understood, such as the user wanting to obtain case handling strategies (settlement, mediation, and court hearings, etc.) or the user wanting to register the judgment result. Next, the user's intent is broken down into sub-tasks, that is, possible execution paths are generated. For example, for the user wanting to obtain litigation case handling strategies, sub-tasks include knowledge query sub-tasks, data query sub-tasks, evidence collection sub-tasks, risk diagnosis sub-tasks, etc.; each sub-task includes instructions and a corresponding task execution module, wherein the instructions include task objectives, context information, execution requirements (how to execute, what tools to use, etc.), output specifications (output format and content), and feedback path (where the result is returned), etc. After receiving the intermediate execution results from the task execution module, the current state is analyzed based on the intermediate execution results to determine the next action. If the current state information is insufficient to support the task planning module in making an initial execution strategy, a sub-task is generated to obtain relevant information. If the current state information is sufficient to support the task planning module in making an initial execution strategy, an initial execution strategy is generated and sent to the strategy output module.

[0044] The task execution module 2 is used to receive the sub-tasks sent by the task planning module 1, execute the sub-tasks to obtain intermediate execution results, and send the intermediate execution results to the task planning module 1.

[0045] The strategy output module 3 is used to process the initial execution strategy sent by the task planning module 1, obtain strategy display information suitable for displaying to the user, and output it.

[0046] The memory module 4 is used to store short-term memory and long-term memory. Among them, short-term memory saves the context of the current session, and long-term memory stores user habits, preferences, historical behaviors, etc. Specifically, historical dialogues, documents, and experiences are encoded into vectors. When a new task arrives, the most relevant memory fragments are retrieved and provided to each agent as context reference.

[0047] In a specific example, when the e-commerce platform is the defendant, it can use the above-mentioned auxiliary case handling system to receive the complaint, plaintiff's evidence, and other materials sent by the court, as well as the case handling rules, to allow the large model to formulate specific case handling strategies and produce planning tasks. For example: a consumer buys an item on the platform, is guided by the merchant to purchase a third-party link, and confirms receipt. After receiving the complaint and the other party's evidence, the task planning module 1 determines that it is necessary to recall the Civil Code for Transaction Disputes and the platform rules, and generates a knowledge query sub-task. The knowledge query module 22 obtains the Civil Code for Transaction Disputes and the platform rules. Then, based on information such as the Civil Code for Transaction Disputes, platform rules, complaints, and plaintiff's evidence, task planning continues. For example, the focus of such cases is whether our side disclosed merchant information in a timely manner. Thus, an evidence collection sub-task is generated, where the evidence collection module 23 obtains our evidence. Simultaneously, the risk diagnosis module 26 can be called in parallel to assess general risks. The task planning module 1 determines whether the litigation case has risks based on our evidence, plaintiff's evidence, case focus, and general risks. Finally, based on the overall risk assessment result, a strategy is generated to create sub-tasks and sent to the delegation module 24. The delegation module 24 then obtains the recommended execution strategy for the case. The task planning module 1 generates an initial execution strategy based on the recommended execution strategy. For example, if customer service has flaws in communication, the risk is high, and a court hearing is suitable. Finally, the initial execution strategy is sent to the strategy output module 3 for summary and user reference.

[0048] In this embodiment of the invention, the task planning module determines the next planning task based on the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module executes the sub-tasks generated by the task planning module to obtain the intermediate execution results and sends the intermediate execution results to the task planning module. The strategy output module obtains strategy display information based on the initial execution strategy generated by the task planning module and outputs it. Therefore, automated auxiliary processing of cases can be achieved, improving the efficiency of litigation case processing, shortening case processing time, and reducing labor costs.

[0049] Figure 3 is a flowchart of the auxiliary case-handling method according to an embodiment of the present invention. As shown in Figure 3, the auxiliary case-handling method according to an embodiment of the present invention includes the following steps: Step S100: Determine the next planning task based on the current status information of the target case.

[0050] In this embodiment, after the task planning module obtains the case information, or after receiving the intermediate execution result returned by the task execution module, it determines the next planning task based on the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating initial execution strategies.

[0051] For the task planning module to determine the next planning task based on the current status information of the target case after obtaining the case information, it includes: receiving case information input by the user, the case information including questions input by the user and / or litigation documents, the litigation documents including documents submitted by the parties and / or documents made by the court. For example, when a user needs to process a case, they input relevant questions through the human-computer interaction interface, such as "Please give the processing strategy for case number xx", "Please give the processing strategy for case name xx", etc. Meanwhile, litigation documents can also be added to the human-computer interaction interface, such as the complaint, the opposing party's evidence, and the judgment (in formats such as PDF and Word); or, text content can be added, that is, the text content in the complaint, the opposing party's evidence, and the judgment can be copied into the user's input question. After the task planning module obtains the case information, it determines the next planning task based on the current status information of the target case. At this time, the current status information includes the case information, and the next planning task is to generate a subtask. The subtask includes an instruction and a target task execution module, wherein the target task execution module is used to indicate which task execution module will execute the subtask, and the instruction is the detailed content of the subtask. The task execution module is used to execute the subtask generated by the task planning module to obtain the intermediate execution result, and send the intermediate execution result to the task planning module.

[0052] More specifically, after the current status information obtains information such as the case information, it obtains short-term memory and long-term memory from the memory module and performs context construction, including merging the user question, short-term memory, long-term memory, and case information. Based on the constructed context, the user's intent is understood. For example, the user may want to obtain information on litigation case handling strategies (settlement, mediation, and court hearings), or the user may want to register the judgment result. Subtasks are generated based on the user's intent, that is, possible execution paths are generated. For example, for the user wanting to obtain information on litigation case handling strategies, subtasks include knowledge query subtasks, data query subtasks, evidence collection subtasks, and risk diagnosis subtasks; each subtask includes instructions and a corresponding task execution module.The block, wherein the instruction includes task objective, context information, execution requirements (how to execute, what tools to use, etc.), output specifications (output format and content), feedback path (where the result is returned), etc.

[0053] For example, for the data query subtask, its corresponding target task execution module is the data query module, and its corresponding instruction on page 6 / 11 of the specification, 10 CN 121563719 A, can be: use the xx tool to search for xx information from the xx database, and return the searched information to the task planning module.

[0054] For the knowledge query subtask, its corresponding target task execution module is the knowledge query module, and its corresponding instruction can be: retrieve xx information from the knowledge base, organize the retrieved information, and return it to the task planning module.

[0055] For the evidence investigation subtask, its corresponding target task execution module is the evidence investigation module, and its corresponding instruction can be: search for xx-related evidence from the evidence base, and return the searched evidence to the task planning module.

[0056] For the strategy generation subtask, the corresponding target task execution module is the delegation module, and the corresponding instruction can be: Please generate a recommended execution strategy based on the information provided below. The recommended execution strategy includes reconciliation, mediation, and court hearing, and return the recommended execution strategy to the task planning module.

[0057] For the registration subtask, the corresponding target task execution module is the registration module, and the corresponding instruction can be: Please register the xx document in the case management system.

[0058] For the risk diagnosis subtask, the corresponding target task execution module is the risk diagnosis module, and the corresponding instruction can be: Please determine whether the case has xx risk based on the information provided below, and return the risk result to the task planning module.

[0059] After receiving the intermediate execution result returned by the task execution module, the task planning module determines the next planning task based on the current status information of the target case, including: The task planning module analyzes and reflects on the case information and intermediate execution results. If the current status information is insufficient to support the task planning module in making an initial execution strategy, then it continues to generate subtasks to obtain relevant information; if the current status information is sufficient to support the task planning module in making an initial execution strategy, then it generates an initial execution strategy and sends the initial execution strategy to the strategy output module.

[0060] It should be noted that the above description uses user input of litigation documents as an example, but the embodiments of the present invention do not limit the specific implementation method of the system obtaining litigation documents. For example, in a scenario with a case management module, the task planning module can obtain litigation documents from the case management module based on the questions input by the user. For example, the questions input by the user include the case identifier of the target case, such as the case name, case number, etc. For example, when a user needs to review a certain litigation case...During processing, relevant questions are input through the human-computer interaction interface, such as "Please provide a processing strategy for case number xx" or "Please provide a processing strategy for case name xx". After receiving the questions input by the user, the task planning module calls the interface tool to extract the litigation documents corresponding to the case identifier from the case management module.

[0061] Figure 4 is a flowchart of determining the next step of the plan according to an embodiment of the present invention. As shown in Figure 4, the step of determining the next planning task based on the current status information of the target case includes the following steps: Step S110: Determine the processing type of the target case based on the case information.

[0062] In this embodiment, the processing type of the target case is determined based on the questions input by the user and the case information in the case information. The processing type includes strategy generation and information registration. For example, when the task planning module detects that the case information includes a complaint and evidence from the opposing party, it indicates that the current stage of the target litigation case is the stage of being sued, and the processing type is more likely to be the execution strategy of planning the target case. At the same time, the user's intention is determined by combining the questions input by the user, and the processing type of the target case is determined by combining the user's intention and the litigation documents in the case information.

[0063] Step S120: In response to the target case's processing type being strategy generation, a knowledge query subtask is generated to obtain normative documents.

[0064] In this embodiment, in response to the target case's processing type being strategy generation, the process of obtaining the target case's execution strategy is divided into multiple subtasks. First, the case type is determined based on the case information. For example, case types include intellectual property cases, transaction cases, etc.; transaction cases include product quality issues, contract performance issues, false advertising fraud, etc.; intellectual property cases include counterfeit products, copyright infringement, etc. The normative documents for processing different types of cases are different. Therefore, a knowledge query subtask is generated based on the case type and sent to the knowledge query module. For example, assuming the case type is a transaction case, the knowledge query subtask includes obtaining transaction-related legal provisions and platform rules in the Civil Procedure Law, where the platform rules include service agreements, dispute resolution rules, management regulations, etc. The knowledge query subtask is sent to the knowledge query module, which obtains the normative document and sends it to the task planning module.

[0065] Step S130: Determine the focus based on the normative document and the case information.

[0066] In this embodiment, the task planning module analyzes the target case based on the normative information and the case information to determine the focus.

[0067] For example, suppose the target case is: a consumer buys an item on the platform, is guided by the merchant to purchase a third-party link, and confirms receipt. After receiving the complaint and the other party's evidence, the task planning module determines that it is necessary to recall the Civil Code for transaction disputes.The platform rules and regulations are used to generate a knowledge query sub-task, which obtains the Civil Code for Transaction Disputes and the platform rules through the knowledge query module. Then, based on the Civil Code for Transaction Disputes, the platform rules, the complaint, the plaintiff's evidence, and other information, the task planning continues, and the focus of such cases is: whether our side has disclosed the merchant information in a timely manner.

[0068] Step S140: Generate an evidence investigation sub-task based on the focus to obtain our side's evidence.

[0069] In this embodiment, after obtaining the focus, an evidence investigation sub-task is generated based on the focus and sent to the evidence investigation module to obtain our side's evidence through the evidence investigation module.

[0070] For example, assuming the focus of the target case is: whether our side has disclosed the merchant information in a timely manner, a relevant evidence investigation sub-task is generated to obtain whether our side has disclosed the merchant information in a timely manner from our side's evidence. The adjustment module sends the obtained evidence to the task planning module.

[0071] Step S150: Generate a risk diagnosis sub-task to obtain the general risk assessment result of the target case.

[0072] In this embodiment, before, after, or simultaneously with obtaining our evidence, a risk diagnosis sub-task is generated and sent to the risk diagnosis module to obtain a general risk assessment result for the target case through the risk diagnosis module.

[0073] For example, risk diagnosis sub-tasks are generated to detect whether there is criminal risk, whether there are flaws in the customer service process, whether the platform promptly leaks relevant information, whether the platform has taken reasonable measures (including but not limited to: product removal or deletion, freezing of deposit, and assistance with refund services), whether the platform has no intentional or gross negligence, and whether the buyer has provided evidence of authenticity, etc., and the sub-tasks are sent to the risk diagnosis module to obtain a general risk assessment result for the target case through the risk diagnosis module. The risk diagnosis module sends the general risk assessment result to the task planning module.

[0074] Step S160: Determine the overall risk assessment result of the target case based on one or more of the case information, focus, our evidence, and general risk assessment result.

[0075] In this embodiment, after obtaining information such as case information, points of concern, our evidence, and general risk assessment results, the task planning module proceeds to the next step. If the current information is complete, the overall risk assessment result of the case is determined based on one or more of the case information, points of concern, our evidence, and general risk assessment results. If the current information is incomplete, sub-tasks are generated based on the current information to continue obtaining intermediate execution results until the information is complete.

[0076] For example, if our evidence shows no fault in the points of concern and no risk is found in the general risk assessment results, the overall risk assessment result can be determined to be risk-free. If our evidence shows fault in the points of concern, or risk is found in the general risk assessment results, the overall risk assessment result can be determined to be risky.

[0077] Step S170: Generate a strategy and sub-task based on the overall risk assessment result to obtain the recommended execution strategy for the target case.

[0078] In this embodiment, the task planning module generates a strategy and sub-task based on the overall risk assessment result and sends it to the delegation module to obtain the recommended execution strategy for the case through the delegation module. The strategy generation sub-task includes information such as the overall risk assessment result and risk types. The delegation module obtains the recommended execution strategy based on the strategy generation sub-task. The recommended execution strategy includes settlement, mediation, and court hearing. The delegation module sends the generated recommended execution strategy to the task planning module.

[0079] Step S180: Generate the initial execution strategy based on the recommended execution strategy.

[0080] In this embodiment, the task planning module obtains the initial execution strategy based on the recommended execution strategy. Specifically, the initial execution strategy is generated based on one or more of the recommended execution strategy, case information, focus, our evidence, general risk assessment result, and overall risk assessment result.

[0081] The initial execution strategy includes a recommended execution strategy and reasons for recommendation. The recommended execution strategy includes settlement, mediation, and court hearing, and the reasons for recommendation are generated based on one or more of the following: case information, points of concern, our evidence, general risk assessment results, and overall risk assessment results.

[0082] Step S190: In response to the processing type of the target case being information registration, an information registration sub-task is generated to enter litigation documents into the case management module.

[0083] Further, in step S110, if the processing type is determined to be information registration, a registration sub-task is generated and sent to the registration module to register the litigation documents through the registration module. The registration module has access to the case management module and can register litigation documents in the case management module. At the same time, the intermediate execution result is returned to the case management module, which can be the registration result, such as registration success or registration failure.

[0084] Step S200: The sub-task is executed to obtain the intermediate execution result.

[0085] In this embodiment, in step S100 above, the task planning module sends the generated subtasks to the corresponding task execution module. The task execution module executes the corresponding subtasks according to the instructions to obtain intermediate execution results, and then sends the intermediate execution results to the task planning module.

[0086] Wherein, executing the subtasks to obtain the intermediate execution results includes at least one of the following: calling the corresponding access tool to obtain relevant data from the data storage module according to the instructions; obtaining the corresponding normative documents from the knowledge storage module according to the instructions; obtaining our evidence according to the instructions; obtaining a recommended execution strategy according to the instructions, the recommended execution strategy including reconciliation, mediation, and court hearing;According to the instructions, enter the litigation documents in the case management module; according to the instructions, obtain the general risk assessment result of the target case.

[0087] Step S300: Obtain and output the strategy display information according to the initial execution strategy.

[0088] In this embodiment, after the information output module processes the initial execution strategy generated by the task planning module, it obtains and outputs the strategy display information of the target case.

[0089] Specifically, Figure 5 is a flowchart of obtaining strategy display information according to an embodiment of the present invention. As shown in Figure 5, obtaining and outputting strategy display information according to the initial execution strategy includes the following steps: Step S310: Convert the initial execution strategy into structured data.

[0090] In this embodiment, the information output module receives the initial execution strategy sent by the task planning module. The initial execution strategy can be structured and / or unstructured information, including one or more of text, tables, code, image descriptions, logs, JSON, etc. Then, the unstructured data (such as logs, code, long text) is converted into structured data. Instruction manual, page 9 / 11, 13 CN 121563719 A. Thus, structured data can be obtained.

[0091] Step S320: Extract key information from the structured data using a large language model.

[0092] In this embodiment, LLM is used to extract key information from the structured data, such as summaries and conclusions, and redundant information is removed.

[0093] Step S330: Generate intermediate content based on the key information.

[0094] In this embodiment, LLM is used to refine the key information, improving readability and professionalism to obtain intermediate content.

[0095] Step S340: Adjust the intermediate content to obtain the strategy display information.

[0096] In this embodiment, the intermediate content is formatted by adding titles, paragraphs, lists, citations, code blocks, bolding, etc., and then the format is adjusted according to the human-computer interaction interface to obtain strategy display information, which is then sent to the human-computer interaction interface for display.

[0097] It should be noted that the task planning module, task execution module, information output module, etc., can all retrieve short-term and long-term memories from the memory module when performing their own tasks, and combine the short-term and long-term memories to execute the corresponding tasks.

[0098] In this embodiment of the invention, the task planning module determines the next planning task based on the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module executes the sub-tasks generated by the task planning module to obtain the intermediate execution results and sends the intermediate execution results to the task planning module.The strategy output module obtains and outputs strategy display information based on the initial execution strategy generated by the task planning module. This enables automated auxiliary processing of cases, improves the efficiency of litigation case handling, shortens case processing time, and reduces labor costs.

[0099] Figure 6 is a schematic diagram of the auxiliary case-handling device according to an embodiment of the present invention. As shown in Figure 6, the auxiliary case-handling device includes a planning unit 71, an execution unit 72, and a display unit 73. The planning unit 71 is used to determine the next planning task based on the current status information of the target case, which includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. The execution unit 72 is used to execute the sub-tasks to obtain the intermediate execution results. The display unit 73 is used to obtain and output strategy display information based on the initial execution strategy.

[0100] In this embodiment of the invention, the task planning module determines the next planning task based on the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module executes the sub-tasks generated by the task planning module to obtain the intermediate execution results and sends the intermediate execution results to the task planning module. The strategy output module obtains strategy display information based on the initial execution strategy generated by the task planning module and outputs it. Therefore, automated auxiliary processing of cases can be achieved, improving the efficiency of litigation case processing, shortening case processing time, and reducing labor costs.

[0101] Figure 7 is a schematic diagram of an electronic device according to an embodiment of the invention. In this embodiment, the electronic device 8 includes a server, a terminal, etc. As shown in Figure 7, the electronic device 8 includes at least one processor 81; a memory 82 communicatively connected to at least one processor 81; and a communication component 83 communicatively connected to a scanning device. The communication component 83 receives and transmits data under the control of the processor 81. The memory 82 stores instructions executable by at least one processor 81, which are executed by the processor 81 to implement the aforementioned auxiliary case-handling method.

[0102] Specifically, the electronic device includes one or more processors 81 and a memory 82. Figure 7 shows an example of one processor 81. The processor 81 and the memory 82 can be connected via a bus or other means. Figure 7 shows an example of connection via a bus. The memory 82, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The processor 81 runs the non-volatile software programs stored in the memory 82.The memory 82 includes a program storage area and a data storage area. The program storage area can store the operating system and at least one application required for a function; the data storage area can store an option list, etc. In addition, the memory 82 can include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 82 may optionally include memory remotely located relative to the processor 81. These remote memories can be connected to external devices via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0104] One or more modules are stored in the memory 82. When executed by one or more processors 81, they execute the auxiliary case-handling method in any of the above method embodiments.

[0105] The above product can execute the method provided in the embodiments of this application, and has the corresponding functional modules and beneficial effects of the method. For technical details not described in detail in this embodiment, please refer to the method provided in the embodiments of this application.

[0106] In this embodiment of the invention, the task planning module determines the next planning task based on the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module executes the sub-tasks generated by the task planning module to obtain the intermediate execution results and sends the intermediate execution results to the task planning module. The strategy output module obtains strategy display information based on the initial execution strategy generated by the task planning module and outputs it. Thus, automated auxiliary processing of cases can be achieved, improving the efficiency of litigation case processing, shortening case processing time, and reducing labor costs.

[0107] Another embodiment of the invention relates to a non-volatile storage medium for storing a computer-readable program, which is used for a computer to execute some or all of the above-described method embodiments.

[0108] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0109] The above description is only a preferred embodiment of this application and is not intended to limit this application. For those skilled in the art, this application can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the protection scope of this application. Instruction Manual 11 / 11 Page 15 CN 121563719 A Figure 1 Instruction Manual Figure 1 / 7 Page 16 CN 121563719 A Figure 2 Instruction Manual Figure 2 / 7 Page 17 CN 121563719 A Figure 3 Instruction Manual Figure 3 / 7 Page 18 CN 121563719 A Figure 4 Instruction Manual Figure 4 / 7 Page 19 CN 121563719 A Figure 5 Instruction Manual Figure 5 / 7 Page 20 CN 121563719 A Figure 6 Instruction Manual Figure 6 / 7 Page 21 CN 121563719 A Figure 7 Instruction Manual Figure 7 / 7 Page 22 CN 121563719 A Abstract Embodiments of the present application disclose a kind of auxiliary case handling system, method and device. By task planning module, next step planning task is determined according to the current state information of target case, the current state information of target case includes case information and / or intermediate execution result, the next step planning task generating includes subtask and / or generating initial execution strategy, at least one task execution module executes the subtask generated by the task planning module to obtain the intermediate execution result, and theintermediate execution result is sent to the task planning module, and strategy output module obtains strategy display information according to the initial execution strategy generated by the task planning module and outputs. Therefore, the automatic auxiliary processing of case can be realized, the processing efficiency of litigation case is improved, the case processing time is shortened, and the labor cost is reduced.

Claims

1. A case-handling auxiliary system, characterized in that, The system includes: The task planning module is used to determine the next planning task based on the current status information of the target case. The current status information of the target case includes case information and / or intermediate execution results. The next planning task includes generating sub-tasks and / or generating an initial execution strategy. At least one task execution module is configured to execute subtasks generated by the task planning module to obtain the intermediate execution results, and send the intermediate execution results to the task planning module; The strategy output module is used to obtain and output strategy display information based on the initial execution strategy generated by the task planning module.

2. The system according to claim 1, characterized in that, The task planning module, at least one task execution module, and the strategy output module are intelligent agents.

3. The system according to claim 2, characterized in that, The system also includes: Memory modules are used to store long-term and short-term memories; The task planning module, at least one task execution module, and strategy output module are used to retrieve the long-term memory and short-term memory from the memory module.

4. The system according to claim 1, characterized in that, The system also includes: The human-computer interaction module is used to receive case information input by the user, including questions input by the user and / or litigation documents, including documents submitted by the parties and / or documents prepared by the court; and / or to receive and display strategy display information sent by the strategy output module.

5. The system according to claim 4, characterized in that, The subtask includes instructions and target task execution modules.

6. The system according to claim 5, characterized in that, The system also includes at least one of the following: The case management module is used to manage all cases. The knowledge storage module is used to store normative documents, which include one or more of the following: laws and regulations, platform rules, and rules and regulations. Data storage module, used to store data; The tool center includes at least one access tool for retrieving data from the data storage module.

7. The system according to claim 6, characterized in that, The task execution module includes at least one of the following: The data query module is used to call the corresponding access tool to obtain relevant data from the data storage module according to the instruction; The knowledge query module is used to retrieve the corresponding normative documents from the knowledge storage module according to the instructions. The evidence collection module is used to obtain our evidence according to the instructions. The delegation module is used to obtain recommended execution strategies based on the instructions, and the recommended execution strategies include settlement, mediation, and court hearing. The registration module is used to enter litigation documents into the case management module according to the instructions. The risk diagnosis module is used to obtain the general risk assessment result of the target case according to the instructions.

8. The system according to claim 7, characterized in that, The task planning module is used to determine the processing type of the target case based on the case information. The processing type includes strategy generation and information registration. In response to the processing type of the target case being strategy generation, a knowledge query subtask is generated and sent to the knowledge query module. The knowledge query module is used to execute the knowledge query subtask to obtain normative documents; The task planning module is used to determine the focus of the target case based on the normative document, and generate investigation sub-tasks based on the focus; The evidence investigation module is used to execute the evidence investigation sub-tasks to obtain our evidence; The task planning module is used to generate risk diagnosis sub-tasks; The risk diagnosis module is used to execute the risk diagnosis sub-task to obtain a general risk assessment result for the target case; The task planning module is used to determine the overall risk assessment result of the target case based on one or more of the case information, focus, our evidence, and general risk assessment results, and to generate sub-tasks based on the overall risk assessment result; The delegation module is used to execute the strategy generation sub-task to obtain the recommended execution strategy for the target case; The task planning module is used to generate the initial execution strategy based on the recommended execution strategy.

9. The system according to claim 8, characterized in that, The task planning module is also used to generate an information registration sub-task and send it to the registration module in response to the target case's processing type being information registration; The registration module is used to perform the information registration sub-task to enter litigation documents into the case management module.

10. The system according to claim 1, characterized in that, The strategy output module is used to convert the initial execution strategy into structured data, extract key information from the structured data using a large language model, generate intermediate content based on the key information, and adjust the intermediate content to obtain the strategy display information.

11. A method for assisting in case handling, characterized in that, The method includes: The next planning task is determined based on the current status information of the target case, wherein the current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating an initial execution strategy; Execute the subtask to obtain the intermediate execution result; The strategy display information is obtained and output according to the initial execution strategy.

12. The method according to claim 11, characterized in that, The method further includes: Receive case information input by the user, including questions input by the user and / or litigation documents, including documents submitted by the parties and / or documents prepared by the court.

13. The method according to claim 11, characterized in that, The subtask includes instructions and target task execution modules.

14. The method according to claim 13, characterized in that, The execution of the subtask to obtain the intermediate execution result includes at least one of the following: The corresponding access tool is invoked according to the instruction to retrieve relevant data from the data storage module; According to the instructions, retrieve the corresponding normative documents from the knowledge storage module; Obtain our evidence according to the instructions; The recommended execution strategy is obtained according to the instruction, and the recommended execution strategy includes settlement, mediation, and court hearing; Enter litigation documents into the case management module according to the instructions; Obtain the general risk assessment result of the target case according to the instructions.

15. The method according to claim 14, characterized in that, The step of determining the next planning task based on the current status information of the target case includes: The processing type of the target case is determined based on the case information, and the processing type includes strategy generation and information registration; In response to the target case's processing type being strategy generation, a knowledge query subtask is generated to obtain normative documents; The focus of the target case is determined based on the aforementioned normative documents; Based on the aforementioned concerns, generate investigation sub-tasks to obtain our evidence; Generate a risk diagnosis subtask to obtain a general risk assessment result for the target case; The overall risk assessment result of the target case is determined based on one or more of the case information, points of concern, our evidence, and general risk assessment results. Based on the overall risk assessment results, a strategy is generated to create sub-tasks to obtain the recommended execution strategy for the target case; The initial execution strategy is generated based on the recommended execution strategy.

16. The method according to claim 15, characterized in that, The step of determining the next planning task based on the current status information of the target case also includes: In response to the target case being processed as information registration, an information registration subtask is generated to enter litigation documents into the case management module.

17. The method according to claim 11, characterized in that, The step of obtaining and outputting strategy display information based on the initial execution strategy includes: Convert the initial execution strategy into structured data; Key information is extracted from the structured data using a large language model; Intermediate content is generated based on the key information; The intermediate content is adjusted to obtain the strategy display information.

18. An auxiliary case-handling device, characterized in that, The device includes: The planning unit is used to determine the next planning task based on the current status information of the target case, wherein the current status information of the target case includes case information and / or intermediate execution results, and the next planning task includes generating sub-tasks and / or generating an initial execution strategy. An execution unit is used to execute the subtasks to obtain the intermediate execution results; The display unit is used to obtain and output strategy display information based on the initial execution strategy.

19. An electronic device comprising a memory and a processor, characterized in that, The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method as described in any one of claims 11-17.

20. A computer-readable storage medium storing computer program instructions thereon, characterized in that, The computer program instructions, when executed by a processor, implement the method as described in any one of claims 11-17.

21. A computer program product, the computer program product comprising a computer program, characterized in that, When the computer program is run on a computer, the computer performs the method according to any one of claims 11-17.

Images