Duty review material auditing method and device, terminal and storage medium

Abstract

The invention provides a title review material auditing method and device, a terminal and a storage medium, and belongs to the technical field of auditing authentication, and the method comprises the steps: receiving a plurality of pieces of review related information and corresponding title review materials; generating a first prompt word according to all the review related information; preprocessing all the title review materials to obtain a plurality of pictures, and inputting all the pictures and a first cue word into a multi-modal large model to obtain a first judgment conclusion; after extracting texts from all the pictures, inputting the texts and the first cue word into the semantic large model to obtain a second judgment conclusion; and based on the first judgment conclusion and the second judgment conclusion, determining a final conclusion about the correlation between the review related information and the title review material, and outputting the final conclusion. According to the invention, the correlation between the review related information and the title review material is judged by using the two models, and the correlation between the review related information and the title review material is determined based on the output results of the two models, so that the review efficiency can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of audit and certification, and particularly relates to a method, device, terminal and storage medium for auditing professional title evaluation materials. Background Art

[0002] In the prior art, an applicant for professional title evaluation needs to select a series of materials for supporting the application on the declaration interface, including papers, patents, projects, awards, etc., and upload the corresponding files. However, at present, there is a lack of effective means to ensure that the information entered by the applicant in the form is exactly the same as the uploaded attachments, and it still relies on manual checking one by one. This manual checking method has low audit efficiency.

[0003] Therefore, there are defects in the prior art and it needs to be improved and developed. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, device, terminal and storage medium for auditing professional title evaluation materials in view of the above-mentioned defects of the prior art, aiming to solve the problem of low audit efficiency in the manual checking method of the prior art.

[0005] The technical solution adopted by the present invention to solve the technical problem is as follows: In a first aspect, an embodiment of the present invention provides a method for auditing professional title evaluation materials, and the method includes: Receiving several pieces of evaluation-related information input by a user on the professional title evaluation system interface, and the corresponding professional title evaluation materials uploaded by the user for each piece of evaluation-related information; Generating a first prompt word according to all the evaluation-related information; Preprocessing all the professional title evaluation materials to obtain several pictures, inputting all the pictures and the first prompt word into a multimodal large model, and after being processed by the multimodal large model, outputting a first judgment conclusion on the relevance between the evaluation-related information and the professional title evaluation materials; After extracting text from all the pictures, inputting the text and the first prompt word into a semantic large model, and after being processed by the semantic large model, outputting a second judgment conclusion on the relevance between the evaluation-related information and the professional title evaluation materials; Based on the first judgment conclusion and the second judgment conclusion, determining and outputting a final conclusion on the relevance between the evaluation-related information and the professional title evaluation materials.

[0006] In an implementation manner, each piece of evaluation-related information includes the type of professional title evaluation materials and the corresponding inspection item content; generating a first prompt word according to all the evaluation-related information includes: Pre-setting inspection items corresponding to each type of professional title evaluation materials, and configuring a corresponding prompt word template for each inspection item; Determine all corresponding inspection items according to the type of the title evaluation materials input by the user; Obtain the prompt word template corresponding to each inspection item, and fill the relevant inspection item content input by the user into the corresponding prompt word template; When all inspection item contents are filled into the corresponding prompt word templates, generate a first prompt word based on the filled prompt word templates.

[0007] In one implementation, preprocess all the title evaluation materials to obtain several pictures, including: Judge the formats of all the title evaluation materials to distinguish picture format files and non-picture format files; For the title evaluation materials in non-picture format, convert them into pictures; For the title evaluation materials in picture format, judge whether their resolution reaches the preset resolution; If the preset resolution is not reached, generate a re-upload instruction including a resolution requirement description, and receive the new pictures submitted by the user in response to the re-upload instruction, and continue to judge whether the resolution of the new pictures reaches the preset resolution until pictures reaching the preset resolution are obtained; The pictures reaching the preset resolution and the pictures after format conversion together form the several pictures.

[0008] In one implementation, for the title evaluation materials in non-picture format, convert them into pictures, including: Judge whether the number of pages of each non-picture format title evaluation material exceeds the preset number of pages; If the number of pages of the non-picture format title evaluation material does not exceed the preset number of pages, perform format conversion on it to obtain the corresponding picture; If the number of pages of the non-picture format title evaluation material exceeds the preset number of pages, intercept it according to the preset number of pages, and perform format conversion on the intercepted non-picture format title evaluation material to obtain the corresponding picture.

[0009] In one implementation, based on the first judgment conclusion and the second judgment conclusion, determine and output the final conclusion on the relevance between the evaluation-related information and the title evaluation materials, including: Judge whether the first judgment conclusion and the second judgment conclusion are consistent; If the two are consistent, use the first judgment conclusion or the second judgment conclusion as the final conclusion and output it; If the two are inconsistent, construct a second prompt word, input the pictures and the second prompt word into the multi-modal large model, and output the final conclusion after processing.

[0010] In one embodiment, while outputting the first judgment conclusion, the multimodal large model outputs a first reasoning process, where the first reasoning process is the process by which the multimodal large model reasons about the relevance between the review-related information and the professional title review materials. While outputting the second judgment conclusion, the semantic large model outputs a second reasoning process, where the second reasoning process is the process by which the semantic large model reasons about the relevance between the review-related information and the professional title review materials.

[0011] In one embodiment, the steps of constructing the second prompt include: Obtain a preset second prompt template, which is used for the multimodal large model to make a binary choice judgment; Fill the first prompt, the first judgment conclusion, the first reasoning process, the second judgment conclusion, and the second reasoning process into the second prompt template to obtain the second prompt.

[0012] In a second aspect, an embodiment of the present invention further provides a professional title review material review device, including: A data receiving module, configured to receive several items of review-related information input by the user on the professional title review system interface, and the corresponding professional title review materials uploaded by the user for each item of review-related information; A prompt generation module, configured to generate a first prompt according to all the review-related information; A first judgment module, configured to preprocess all the professional title review materials to obtain several pictures, input all the pictures and the first prompt into the multimodal large model, and after being processed by the multimodal large model, output a first judgment conclusion on the relevance between the review-related information and the professional title review materials; A second judgment module, configured to input the text and the first prompt into the semantic large model after extracting the text from all the pictures, and after being processed by the semantic large model, output a second judgment conclusion on the relevance between the review-related information and the professional title review materials; A final conclusion generation module, configured to determine and output a final conclusion on the relevance between the review-related information and the professional title review materials based on the first judgment conclusion and the second judgment conclusion.

[0013] In a third aspect, an embodiment of the present invention further provides a terminal, where the terminal includes: a memory, a processor, and a professional title review material review program stored on the memory and executable on the processor. When the professional title review material review program is executed by the processor, the steps of the professional title review material review method described above are implemented.

[0014] Fourthly, an embodiment of the present invention further provides a computer-readable storage medium storing a title evaluation material review program, which can be executed to implement the steps of the title evaluation material review method as described above.

[0015] Advantages of the present invention: The present invention receives a number of review-related information and corresponding title evaluation materials; generates a first prompt word according to all the review-related information; preprocesses all the title evaluation materials to obtain a number of pictures, inputs all the pictures and the first prompt word into a multimodal large model to obtain a first judgment conclusion; after extracting text from all the pictures, inputs the text and the first prompt word into a semantic large model to obtain a second judgment conclusion; based on the first judgment conclusion and the second judgment conclusion, determines and outputs a final conclusion on the relevance between the review-related information and the title evaluation materials. The present invention respectively uses two models to judge the relevance between the review-related information and the title evaluation materials, and determines the relevance between the review-related information and the title evaluation materials based on the output results of the two models, which can effectively improve the review efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flowchart of a preferred embodiment of the title evaluation material review method in the present invention.

[0017] Figure 2 is a data processing flowchart in the present invention.

[0018] Figure 3 is a schematic structural diagram of a preferred embodiment of the title evaluation material review device in the present invention.

[0019] Figure 4 is a principle block diagram of the terminal in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] To make the objectives, technical solutions and advantages of the present invention clearer and more definite, the following further describes the present invention in detail with reference to the accompanying drawings and by way of examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0021] In the prior art, an applicant for title evaluation needs to select a series of materials for supporting the application, such as papers, patents, projects, awards, etc. on the application interface and upload the corresponding files for review. However, at present, there is a lack of effective means to ensure that the information entered by the applicant in the form is completely consistent with the uploaded attachments, and it still relies on manual checking one by one. This manual checking method has low review efficiency.

[0022] In view of the above defects of the prior art, the present invention provides a method, device, terminal and storage medium for reviewing and verifying professional title evaluation materials, belonging to the technical field of review and authentication. The method includes: receiving a number of pieces of review-related information and corresponding professional title evaluation materials; generating a first prompt word according to all the review-related information; preprocessing all the professional title evaluation materials to obtain a number of pictures, and inputting all the pictures and the first prompt word into a multimodal large model to obtain a first judgment conclusion; after extracting text from all the pictures, inputting the text and the first prompt word into a semantic large model to obtain a second judgment conclusion; determining and outputting a final conclusion on the relevance between the review-related information and the professional title evaluation materials based on the first judgment conclusion and the second judgment conclusion. By respectively using two models to judge the relevance between the review-related information and the professional title evaluation materials, and determining the relevance between the review-related information and the professional title evaluation materials based on the output results of the two models, the present invention can effectively improve the review efficiency.

[0023] Please refer to Figure 1 , the method for reviewing and verifying professional title evaluation materials according to the embodiment of the present invention includes the following steps: Step S100, receiving a number of pieces of review-related information input by the user on the interface of the professional title evaluation system, and the corresponding professional title evaluation materials uploaded by the user for each piece of review-related information.

[0024] Specifically, when the user is conducting a professional title evaluation, they need to manually input different review-related information on the professional title evaluation interface and upload the corresponding professional title evaluation materials related to the review-related information.

[0025] Please refer to Figure 1 , the method for reviewing and verifying professional title evaluation materials according to the embodiment of the present invention further includes the following steps: Step S200, generating a first prompt word according to all the review-related information.

[0026] Specifically, the first prompt word can inform the multimodal large model and the language large model of the review-related information that needs to be concerned, guiding the model to specifically extract the content related to the review from the numerous information contained in the pictures or texts and judge its relevance.

[0027] In one implementation, each piece of review-related information includes the type of professional title evaluation materials and the corresponding inspection item content; the generating of the first prompt word according to all the review-related information includes: Pre-setting the inspection items corresponding to each type of professional title evaluation materials, and configuring a corresponding prompt word template for each inspection item; Determining all the inspection items corresponding to it according to the type of professional title evaluation materials input by the user; Obtaining the prompt word template corresponding to each inspection item, and filling the relevant inspection item content input by the user into the corresponding prompt word template; When all the content of the inspection items is filled into the corresponding prompt template, a first prompt is generated based on the filled prompt template.

[0028] Specifically, each item of review-related information includes the type of professional title review materials and the corresponding content of the inspection items. By pre-establishing the correspondence between the type of professional title review materials and the corresponding inspection items, when the user selects the type of review materials on the professional title review system interface, the corresponding inspection items will be automatically brought out on the professional title review system interface. The user can enter the corresponding content of the inspection items in the input box for the content of the inspection items. For example, if the type of professional title review materials is a paper, the corresponding inspection items are the paper title, the name of the journal where the paper is published, the level of the journal where the paper is published, the year when the paper is published, the research field of the paper, the number of citations, and the author information, which the user can fill in manually. In addition, a corresponding prompt template is configured for each inspection item. For example, the prompt template corresponding to the paper title is: "Please judge whether 【】 is the title of the paper in the picture". The content in 【】 is the content that can be replaced by the module. When the user enters the paper title as "Scene Prediction Method Based on Priori Distribution" on the professional title review system interface and fills it into the corresponding prompt template, the prompt template "Please judge whether

Scene Prediction Method Based on Priori Distribution

[0029] Please refer to Figure 1 , the method for reviewing professional title review materials according to the embodiment of the present invention further includes the following steps: Step S300: Preprocess all the professional title review materials to obtain a number of pictures, input all the pictures and the first prompt into a multimodal large model, and after being processed by the multimodal large model, output a first judgment conclusion on the relevance between the review-related information and the professional title review materials.

[0030] Specifically, the present invention uniformly processes different types of professional title review materials into picture form, enabling the multimodal large model to process the materials in a relatively consistent manner. A multimodal large model refers to an artificial intelligence model that can process multiple modalities of information. Modality refers to the source or form of information, such as text, image, audio, video, etc. The present invention can call its processing ability for pictures through the first prompt.

[0031] In one implementation, the multimodal large model is any one of GPT-4o and Tongyi Qianwen model.

[0032] In one implementation, preprocessing all the professional title review materials to obtain a number of pictures includes: Judging the format of all the professional title review materials to distinguish between picture format files and non-picture format files; For the professional title evaluation materials in non - picture format, convert them into pictures; For the professional title evaluation materials in picture format, determine whether their resolution reaches the preset resolution; If the preset resolution is not reached, generate a re - upload instruction containing an explanation of the resolution requirements, and receive the new pictures submitted by the user in response to the re - upload instruction, and continue to determine whether the resolution of the new pictures reaches the preset resolution until pictures with the preset resolution are obtained; Combine the pictures that reach the preset resolution and the pictures after format conversion to form the several pictures.

[0033] Specifically, the professional title evaluation materials uploaded by the user include PDF format and picture format. After receiving the professional title evaluation materials uploaded by the user, first perform format judgment, and convert the non - picture - format professional title evaluation materials (i.e., PDF - format professional title evaluation materials) into pictures. For the professional title evaluation materials that are already in picture format, determine whether the resolution reaches the preset resolution. Only when the preset resolution is reached can the subsequent processing continue. By performing format judgment on the professional title evaluation materials in picture format, the accuracy of information extraction in subsequent professional title evaluations can be effectively ensured, thereby ensuring the accuracy of the final conclusion.

[0034] In one implementation, converting the non - picture - format professional title evaluation materials into pictures includes: Determine whether the number of pages of each non - picture - format professional title evaluation material exceeds the preset number of pages; If the number of pages of the non - picture - format professional title evaluation material does not exceed the preset number of pages, perform format conversion on it to obtain the corresponding pictures; If the number of pages of the non - picture - format professional title evaluation material exceeds the preset number of pages, intercept it according to the preset number of pages, and perform format conversion on the intercepted non - picture - format professional title evaluation material to obtain the corresponding pictures.

[0035] Specifically, in order to effectively control the data volume input into the multi - modal large model, a preset number of pages is set. Materials with more than the preset number of pages are intercepted. In this way, the processing speed of the model can be guaranteed.

[0036] In one implementation, the value range of the preset number of pages is 5 - 30 pages.

[0037] Please refer to Figure 1 , the method for auditing professional title evaluation materials described in the embodiments of the present invention further includes the following steps: Step S400: After extracting text from all the pictures, input the text and the first prompt word into the semantic large model, and the semantic large model processes and outputs a second judgment conclusion on the relevance between the evaluation - related information and the professional title evaluation materials.

[0038] Specifically, in addition to using the multimodal large language model to judge the relevance of the review-related information and the title review materials, the present invention also uses the language large model for judgment. After extracting text from all the pictures using optical character recognition (OCR) technology, the text and the first prompt word are input into the semantic large model for judgment. The semantic large model is an artificial intelligence model that focuses on understanding and generating semantic information in natural language. By learning a large amount of language data such as text, it can dig out the semantic relationship between words, sentences, and paragraphs in the language, thereby realizing many natural language processing tasks such as text classification, sentiment analysis, question-answering systems, and machine translation.

[0039] In one implementation, the semantic big model is any one of the BERT model (Bidirectional Encoder Representations from Transformers), the ERNIE (Enhanced Representation through kNowledge IntEgration) model, and the XLNet model.

[0040] See also Figure 1 The method for reviewing professional title evaluation materials according to the embodiment of the present invention further comprises the following steps: Step S500: Based on the first judgment conclusion and the second judgment conclusion, determine and output the final conclusion on the relevance of the review-related information and the professional title review materials.

[0041] Specifically, optical character recognition (OCR) technology can extract text from images. However, for professional title review materials, if only a single optical character recognition (OCR) technology is used to extract text from images, it cannot process the original style and order of the materials well, resulting in inaccurate information extraction. If only a single multimodal model or semantic large model is used for processing, since its processing method is based on different focuses, it also has one-sided judgment. The present invention combines the judgment conclusions output by the two models to determine and output the final conclusion on the relevance of review-related information and professional title review materials. This method can effectively utilize the advantages of the two models and make the final conclusion more accurate. Specifically, it is determined whether the first judgment conclusion and the second judgment conclusion are consistent; if the two are consistent, the first judgment conclusion or the second judgment conclusion is used as the final conclusion and output. Due to the different focuses of the two models, the multimodal large model can comprehensively consider the text, graphics, format and other information in the picture, and use its ability to perceive the image as a whole to judge the relevance. The semantic big model extracts the text based on the picture and then combines it with the first prompt word to draw the second judgment conclusion, focusing on the in-depth understanding of the text semantics. When the two judgment conclusions are consistent, the present invention uses the first judgment conclusion or the second judgment conclusion as the final conclusion and outputs it, which can effectively avoid the problem of inaccurate judgment. If the two are inconsistent, a second prompt word is constructed, and the picture and the second prompt word are input into the multimodal big model, and the final conclusion is output after processing. This move can further mine the picture information, make the final conclusion more in line with the actual situation, and enhance the adaptability to complex and diverse professional title evaluation materials.

[0042] In one implementation, the multimodal big model outputs a first reasoning process while outputting a first judgment conclusion, and the first reasoning process is a process in which the multimodal big model infers the relevance between review-related information and professional title review materials. The semantic big model outputs a second reasoning process while outputting a second judgment conclusion, and the second reasoning process is a process in which the semantic big model infers the relevance between review-related information and professional title review materials.

[0043] Specifically, the multimodal large model can output the first judgment conclusion and the first reasoning process at the same time, and the semantic large model can output the second judgment conclusion and the second reasoning process at the same time.

[0044] In one implementation, the step of constructing the second prompt word includes: Obtaining a preset second prompt word template, where the second prompt word template is used for the multimodal large model to perform a one-or-two judgment; Fill the first prompt, the first judgment conclusion, the first reasoning process, the second judgment conclusion, and the second reasoning process into the second prompt template to obtain the second prompt.

[0045] Specifically, fill the first prompt, the first judgment conclusion, the first reasoning process, the second judgment conclusion, and the second reasoning process into the second prompt template. This allows the multi-modal large model to fully consider all relevant information from before when making a re-judgment, and then make a binary choice, selecting either the first judgment conclusion or the second judgment conclusion as the final conclusion and outputting it. This helps improve the quality of the final conclusion, making the relevance judgment more accurate and reasonable, and better able to handle information conflicts and uncertainties when dealing with complex review materials and other situations.

[0046] The data processing flow chart of the present invention is as Figure 2 shown. The user first enters the type of form input materials and the content of the inspection items, as well as uploads the corresponding professional title review materials on the professional title review system interface. The first prompt can be generated according to the input content. The professional title review materials are preprocessed into several pictures. The pictures and the first prompt are input into the multi-modal large model to obtain the first judgment result and the first reasoning process. After the pictures are subjected to optical character recognition (OCR), text is obtained. The text and the first prompt are input into the semantic large model to obtain the second judgment conclusion and the second reasoning process. Determine whether the two judgment results are consistent. If they are consistent, either judgment result is used as the final result. If they are inconsistent, a second prompt is constructed, and the pictures and the second prompt are input into the multi-modal large model for a binary choice judgment to output the final conclusion.

[0047] In summary, the present invention receives several items of review-related information and the corresponding professional title review materials; generates the first prompt according to all the review-related information; preprocesses all the professional title review materials to obtain several pictures, inputs all the pictures and the first prompt into the multi-modal large model to obtain the first judgment conclusion; after extracting text from all the pictures, inputs the text and the first prompt into the semantic large model to obtain the second judgment conclusion; based on the first judgment conclusion and the second judgment conclusion, determines and outputs the final conclusion regarding the relevance between the review-related information and the professional title review materials. The present invention can effectively improve the review efficiency by separately using two models to judge the relevance between the review-related information and the professional title review materials and determining the relevance between the review-related information and the professional title review materials based on the output results of the two models.

[0048] In one embodiment, as Figure 3 shown, based on the above professional title review material review method, the present invention also correspondingly provides a professional title review material review device, including: A data receiving module 100, configured to receive several pieces of review-related information input by a user on the interface of the professional title review system, as well as the corresponding professional title review materials uploaded by the user for each piece of review-related information; A prompt word generation module 200, configured to generate a first prompt word according to all the review-related information; A first judgment module 300, configured to preprocess all the professional title review materials to obtain several pictures, input all the pictures and the first prompt word into a multimodal large model, and after being processed by the multimodal large model, output a first judgment conclusion on the relevance between the review-related information and the professional title review materials; A second judgment module 400, configured to, after extracting text from all the pictures, input the text and the first prompt word into a semantic large model, and after being processed by the semantic large model, output a second judgment conclusion on the relevance between the review-related information and the professional title review materials; A final conclusion generation module 500, configured to determine and output a final conclusion on the relevance between the review-related information and the professional title review materials based on the first judgment conclusion and the second judgment conclusion.

[0049] In one embodiment, each piece of review-related information includes the type of professional title review materials and the corresponding inspection item content; the prompt word generation module includes: A preset unit, configured to preset the inspection items corresponding to each type of professional title review materials, and configure a corresponding prompt word template for each inspection item; An inspection item determination unit, configured to determine all the inspection items corresponding to the type of professional title review materials input by the user; An information filling unit, configured to obtain the prompt word template corresponding to each inspection item, and fill the relevant inspection item content input by the user into the corresponding prompt word template; An information integration unit, configured to generate a first prompt word based on the filled prompt word templates when all the inspection item contents are filled into the corresponding prompt word templates.

[0050] In one embodiment, the device further includes: A format judgment unit, configured to judge the formats of all the professional title review materials, and distinguish between picture format files and non-picture format files; A picture conversion unit, configured to convert the non-picture format professional title review materials into pictures; A resolution judgment unit, configured to judge whether the resolution of the picture format professional title review materials reaches a preset resolution; A re-upload unit, configured to generate a re-upload instruction including a resolution requirement description if the preset resolution is not reached, receive a new picture submitted by the user in response to the re-upload instruction, and continue to determine whether the resolution of the new picture reaches the preset resolution until a picture reaching the preset resolution is obtained; A plurality of picture generation units, configured to jointly form the plurality of pictures with the pictures reaching the preset resolution and the pictures after format conversion.

[0051] In one embodiment, the device further includes: A first judgment unit, configured to judge whether the number of pages of each of the professional title evaluation materials in a non-picture format exceeds a preset number of pages; A first format conversion unit, configured to perform format conversion on the professional title evaluation materials in a non-picture format if the number of pages thereof does not exceed the preset number of pages, to obtain corresponding pictures; A second format conversion unit, configured to intercept the professional title evaluation materials in a non-picture format according to the preset number of pages if the number of pages thereof exceeds the preset number of pages, and perform format conversion on the intercepted non-picture format professional title evaluation materials, to obtain corresponding pictures.

[0052] In one embodiment, the final conclusion generation module includes: A second judgment unit, configured to judge whether the first judgment conclusion and the second judgment conclusion are consistent; A first conclusion output unit, configured to, if they are consistent, use the first judgment conclusion or the second judgment conclusion as the final conclusion and output it; A second conclusion output unit, configured to, if they are inconsistent, construct a second prompt word, input the picture and the second prompt word into the multi-modal large model, and output the final conclusion after processing.

[0053] In one embodiment, the device further includes: A template acquisition unit, configured to acquire a preset second prompt word template, which is used for the multi-modal large model to make a binary choice judgment; A second prompt word generation unit, configured to fill the first prompt word, the first judgment conclusion, the first reasoning process, the second judgment conclusion and the second reasoning process into the second prompt word template, to obtain a second prompt word.

[0054] Based on the above embodiments, the present invention further provides a terminal, and its principle block diagram can be as Figure 4As shown in the figure. The above terminal includes a processor, a memory, a network interface, and a display screen connected by a device bus. Among them, the processor of the terminal is used to provide computing and control capabilities. The memory of the terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a title evaluation material review program. The internal memory provides an environment for the operation of the operating system and the title evaluation material review program in the non-volatile storage medium. The network interface of the terminal is used to communicate with an external terminal through a network connection. When the title evaluation material review program is executed by the processor, it implements the steps of any one of the above title evaluation material review methods. The display screen of the terminal can be a liquid crystal display screen or an electronic ink display screen.

[0055] Those skilled in the art can understand that Figure 4 the principle block diagram shown in the figure is only a block diagram of some structures related to the solution of the present invention, and does not constitute a limitation on the terminal to which the solution of the present invention is applied. The specific terminal may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0056] In one embodiment, a terminal is provided. The above terminal includes a memory, a processor, and a title evaluation material review program stored on the above memory and executable on the above processor. When the above title evaluation material review program is executed by the above processor, it implements the steps of any one of the title evaluation material review methods provided by the embodiments of the present invention.

[0057] The embodiments of the present invention also provide a computer-readable storage medium. The above computer-readable storage medium stores a title evaluation material review program. When the above title evaluation material review program is executed by a processor, it implements the steps of any one of the title evaluation material review methods provided by the embodiments of the present invention.

[0058] It should be understood that the sequence numbers of the above steps do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0059] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is used as an example for illustration. In practical applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the above-mentioned device can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present invention. The specific working processes of the units and modules in the above-mentioned device can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0060] In the above embodiments, the descriptions of the respective embodiments have their own focuses. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0061] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0062] In the embodiments provided by the present invention, it should be understood that the disclosed device / terminal device and method can be implemented in other ways. For example, the above-described device / terminal device embodiments are only illustrative. For example, the above-mentioned division of modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0063] The above-mentioned embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not deviate from the spirit and scope of the technical solutions of the present invention in essence and should all be included in the protection scope of the present invention.

Claims

1. A method for reviewing materials for professional title evaluation, characterized in that The method includes: Receiving several pieces of review-related information input by the user on the interface of the professional title review system, and the corresponding professional title review materials uploaded by the user for each piece of review-related information; Generating a first prompt word according to all the review-related information; Preprocessing all the professional title review materials to obtain several pictures, inputting all the pictures and the first prompt word into a multimodal large model, and after being processed by the multimodal large model, outputting a first judgment conclusion on the relevance between the review-related information and the professional title review materials; After extracting text from all the pictures, inputting the text and the first prompt word into a semantic large model, and after being processed by the semantic large model, outputting a second judgment conclusion on the relevance between the review-related information and the professional title review materials; Based on the first judgment conclusion and the second judgment conclusion, determining and outputting a final conclusion on the relevance between the review-related information and the professional title review materials.

2. The method for reviewing title evaluation materials according to claim 1, wherein Each piece of review-related information includes the type of professional title review materials and the corresponding inspection item content; Generating a first prompt word according to all the review-related information includes: Presetting the inspection items corresponding to each type of professional title review materials in advance, and configuring a corresponding prompt word template for each inspection item; Determining all the inspection items corresponding to the type of professional title review materials input by the user; Obtaining the prompt word template corresponding to each inspection item, and filling the relevant inspection item content input by the user into the corresponding prompt word template; When all the inspection item contents are filled into the corresponding prompt word templates, generating a first prompt word based on the filled prompt word templates.

3. The method for reviewing professional title evaluation materials according to claim 1, characterized in that Preprocessing all the professional title review materials to obtain several pictures, including: Judging the formats of all the professional title review materials to distinguish picture format files and non-picture format files; For non-picture format professional title review materials, converting them into pictures; For picture format professional title review materials, judging whether their resolution reaches the preset resolution; If the preset resolution is not reached, generating a re-upload instruction including a resolution requirement description, and receiving new pictures submitted by the user in response to the re-upload instruction, and continuing to judge whether the resolution of the new pictures reaches the preset resolution until pictures with the preset resolution are obtained; Combining the pictures with the preset resolution and the pictures after format conversion to form the several pictures.

4. The method for reviewing professional title evaluation materials according to claim 3, characterized in that For non-picture format professional title review materials, converting them into pictures, including: Judging whether the number of pages of each non-picture format professional title review material exceeds the preset number of pages; If the number of pages of the non-picture format professional title review material does not exceed the preset number of pages, converting its format to obtain the corresponding picture; If the number of pages of the non-picture format professional title review material exceeds the preset number of pages, intercepting it according to the preset number of pages, and converting the intercepted non-picture format professional title review material into the corresponding picture.

5. The method for reviewing title evaluation materials according to claim 1, characterized in that, Based on the first judgment conclusion and the second judgment conclusion, determining and outputting a final conclusion on the relevance between the review-related information and the professional title review materials, including: Judging whether the first judgment conclusion and the second judgment conclusion are consistent; If the two are consistent, then use the first judgment conclusion or the second judgment conclusion as the final conclusion and output it; If the two are inconsistent, then construct a second prompt, input the picture and the second prompt into the multimodal large model, and output the final conclusion after processing.

6. The method for reviewing professional title evaluation materials according to claim 5, characterized in that When the multimodal large model outputs the first judgment conclusion, it also outputs the first reasoning process, and the first reasoning process is the process of the multimodal large model reasoning about the relevance between the review-related information and the professional title review materials. When the semantic large model outputs the second judgment conclusion, it also outputs the second reasoning process, and the second reasoning process is the process of the semantic large model reasoning about the relevance between the review-related information and the professional title review materials.

7. The method for reviewing professional title evaluation materials according to claim 6, characterized in that, The steps of constructing the second prompt include: Obtain a preset second prompt template, which is used for the multimodal large model to make a binary choice judgment; Fill the first prompt, the first judgment conclusion, the first reasoning process, the second judgment conclusion, and the second reasoning process into the second prompt template to obtain the second prompt.

8. An apparatus for reviewing materials for professional title evaluation, characterized in that, including: A data receiving module, configured to receive several items of review-related information input by the user on the professional title review system interface, and the corresponding professional title review materials uploaded by the user for each item of review-related information; A prompt generating module, configured to generate a first prompt according to all the review-related information; A first judgment module, configured to preprocess all the professional title review materials to obtain several pictures, input all the pictures and the first prompt into the multimodal large model, and after being processed by the multimodal large model, output a first judgment conclusion on the relevance between the review-related information and the professional title review materials; A second judgment module, configured to extract text from all the pictures, input the text and the first prompt into the semantic large model, and after being processed by the semantic large model, output a second judgment conclusion on the relevance between the review-related information and the professional title review materials; A final conclusion generating module, configured to determine and output a final conclusion on the relevance between the review-related information and the professional title review materials based on the first judgment conclusion and the second judgment conclusion.

9. A terminal, characterized in that, The terminal includes a memory, a processor, and a professional title review material review program stored on the memory and executable on the processor. When the professional title review material review program is executed by the processor, the steps of the professional title review material review method according to any one of claims 1-7 are implemented.

10. A computer-readable storage medium, characterized in that, A professional title review material review program is stored on the computer-readable storage medium. When the professional title review material review program is executed by the processor, the steps of the professional title review material review method according to any one of claims 1-7 are implemented.

Images