Automatic resume generation system based on large language model

Through the automated resume generation system with a large language model, the problem of lack of intelligence and personalization of resume generation tools is solved, and efficient and professional resume generation is achieved to meet the needs of the modern job search market.

CN120542387AInactive Publication Date: 2025-08-26TIANJIN TIANJI ARTIFICIAL INTELLIGENCE CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202510523023.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-24
Publication Date
2025-08-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing resume generation tools lack the ability to process intelligently and recommend personalized, resulting in difficulty in information organization, unprofessional format, insufficient personalization and inefficient efficiency, and cannot meet the needs of the modern job search market.

Method used

An automated resume generation system based on a large language model is adopted, including an information collection module, a data processing module and a resume generation module. User information is obtained through page filling, data analysis, AI dialogue, etc., and data processing and resume generation are used to ensure that the generated resume meets the job needs and is professional and personalized.

Benefits of technology

Significantly shorten the time for resume writing, generate efficient, personalized and professional resumes, improve user experience, and meet the needs of the modern job search market.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120542387A_ABST
    Figure CN120542387A_ABST
Patent Text Reader

Abstract

The invention discloses an automatic resume generation system based on a large language model. The automatic resume generation system comprises an information collection module, a data processing module and a resume generation module. After a user enters the system, the user information JSON is obtained through the information collection module; and the user information is transmitted to the data processing module to optimize and generate a text in the user information, so that the text better meets the requirements of the application posts of the user. And in the data processing module, auxiliary information is generated according to the post applied by the user, a subjective description text in the user information JSON is optimized by using a large language model, and the optimized user information JSON is obtained. And the resume generation module is used for generating a user resume in a Markdwon format by using the optimized user information JSON based on a resume template in a database selected by the user. And after checking and correcting the initial version of the user resume, generating a final pdf format resume. The system has the advantages of being efficient, personalized, simple, easy to use and professional.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of artificial intelligence and natural language processing, and specifically relates to an automatic resume generation system based on a large language model. Background Art

[0002] In today's job market, resumes are an important tool for job seekers to showcase their personal abilities, experience, and skills. However, many job seekers face the following problems when writing their resumes: Difficulty in organizing information: Job seekers find it difficult to effectively organize their scattered personal experiences and skills into resume content that meets recruitment requirements.

[0003] Unprofessional format: Manually written resumes often have problems such as inconsistent formatting and unsightly layout, which affects the recruiter's reading experience.

[0004] Lack of personalization: Resumes for different positions are poorly customized and fail to highlight experiences and skills that are highly relevant to the target position.

[0005] Inefficiency: The resume writing process is time-consuming and labor-intensive, especially for job seekers who need to apply for multiple positions frequently.

[0006] Existing resume generation tools mostly rely on fixed templates and manual input, lacking intelligent processing and personalized recommendation capabilities, and are unable to meet the needs of the modern job market. Therefore, there is an urgent need for an automated resume generation system based on artificial intelligence technology that can quickly generate high-quality, personalized resumes through intelligent dialogue, data analysis, and large language model optimization. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide an automatic resume generation system based on a large language model, which solves the problem in the prior art that resumes cannot meet the needs of the modern job market.

[0008] The present invention adopts the following technical solutions to solve the above technical problems: The automatic resume generation system based on the large language model includes an information collection module, a data processing module, and a resume generation module; The information collection module includes page filling, data parsing, and AI dialogue. Users manually enter basic information in the page filling part as the source data for establishment and generation. The data parsing part extracts relevant content from the documents uploaded by users to fill in resume information. The AI ​​dialogue part extracts relevant content entered by users in the form of chat as resume filling information. The data obtained from these three parts are integrated into the user information database. The data processing module first applies a large language model to generate job information and resume writing guidance text based on the user's job application. Then, based on the user's job application, job information, and resume writing guidance, the large language model is used to expand and optimize the text in the user information database, and the user information database is updated. The resume generation module converts the updated user information format, selects a resume template, applies the large language model to fill the converted user information into the corresponding position of the empty resume template, generates a preliminary version of the user resume, and checks and corrects the preliminary version of the user resume before generating the final PDF format resume.

[0009] The page filling section provides multiple resume field input items. Users manually fill in basic information and save it as source data for resume generation.

[0010] The data parsing component extracts relevant content from uploaded PDF documents to populate resume information. Using a combination of OCR and large language model analysis, data parsing automatically parses any uploaded document and extracts relevant information suitable for resume completion. OCR scans the text, while the large language model integrates, rewrites, and categorizes the initial text. Once categorized, the large language model stores the acquired information in the corresponding sections of a standardized database, serving as the source data for resume generation.

[0011] For the AI ​​dialogue part, we chose to use an instant dialogue system with a large language model to conduct conversations, guiding users to collect the various information required to fill out their resumes in a chat manner. In view of the different experiences of different users and the different orders in which information is filled out, the large language model adopts open-ended question-answering processing.

[0012] The chat interface of the AI ​​dialogue section supports user voice input.

[0013] The user information in the user information database is in JSON format. The data processing module expands and optimizes the text in JSON format. After optimization, the optimized JSON format user information replaces the original user information.

[0014] The resume generation module first applies the large language model to convert the optimized JSON format user information into Markdown format user information. Secondly, it selects the empty resume template and its corresponding CSS configuration file. Then, after completing the format conversion and template selection, it applies the large language model to fill the content in the user information into the corresponding position of the empty resume template, generates a formatted first version of the user resume, and checks the format of the first version of the user resume. Correct any formatting issues in the first version of the user resume, generates a final version of the user resume, and finally generates a resume in PDF format.

[0015] Before the resume generation operation, several empty resume templates in Markdown format and their corresponding CSS configuration files are pre-acquired and stored.

[0016] The final version of the user resume is written in Markdown syntax and is equipped with a corresponding CSS file.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. This solution is efficient and personalized. It significantly shortens resume writing time through automated information collection and processing. It also generates highly customized resume content through intelligent recommendation and optimization based on a large language model.

[0018] 2. It can automatically generate high-quality personalized resumes through multimodal data input (such as text, documents, and voice) and intelligent dialogue interaction, reducing the difficulty of user operation.

[0019] 3. Professionalism: Automatically generate a resume format that meets industry standards to improve the professionalism and readability of your resume. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of the system of the present invention.

[0021] Figure 2 This is a schematic diagram of the information collection module of the present invention.

[0022] Figure 3 Schematic diagram of the data processing module of the present invention.

[0023] Figure 4 This is a schematic diagram of the resume generation module of the present invention. DETAILED DESCRIPTION

[0024] In order to enable people in this technical field to better understand the solution of this application, the technical solution in the embodiment of this application will be clearly and completely described below in combination with the drawings in the embodiment of this application. Obviously, the described embodiment is only an embodiment of a part of this application, not all embodiments.

[0025] Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work should fall within the scope of protection of this application.

[0026] The automatic resume generation system based on the large language model includes an information collection module, a data processing module, and a resume generation module; The information collection module includes page filling, data parsing, and AI dialogue. Users manually enter basic information in the page filling part, which serves as the source data for resume generation. The data parsing part extracts relevant content from the documents uploaded by the user to fill in the resume information. The AI ​​dialogue part extracts relevant content entered by the user in the form of chat as the resume filling information. The data obtained from these three parts are integrated into the user information database. The data processing module first applies a large language model to generate job information and resume writing guidance text based on the user's job application. Then, based on the user's job application, job information, and resume writing guidance, the large language model is used to expand and optimize the text in the user information database, and the user information database is updated. The resume generation module converts the updated user information format, selects a resume template, applies the large language model to fill the converted user information into the corresponding position of the empty resume template, generates a preliminary version of the user resume, and checks and corrects the preliminary version of the user resume before generating the final PDF format resume.

[0027] Specific embodiments, such as Figures 1 to 4 As shown, The present invention proposes an innovative automatic resume generation system based on a large language model, which consists of three modules: information collection, data processing, and resume generation.

[0028] Among them, the information collection module specifically includes: 1.1 User filling: The system provides users with multiple resume field input items. Users can manually fill in basic information such as name, educational background, work experience, etc. This information will be used as source data for resume generation.

[0029] 1.2 PDF Parsing: The system can parse PDF documents uploaded by users (such as existing resumes, certificates, etc.) and extract relevant content to fill in resume information. Due to the difficulty that the format and content of different users' existing resumes and documents are different, this system uses a combination of OCR and large-scale model analysis to automatically parse any document uploaded by the user and extract relevant information suitable for resume filling. Among them, OCR is responsible for scanning text content, while the large-scale model will integrate, rewrite, and classify the initially acquired text. After the classification is completed, the large-scale model will store the acquired information in a standardized database as the source data for resume generation.

[0030] 1.3 AI Conversation: Users can choose to freely engage in conversations with AI using the large language model's instant conversation system. After initiating the conversation, the large language model will guide users through a chat to collect the various information required to complete a resume, such as personal information, educational background, and project experience. To account for the varying experiences and order in which information is entered, the large language model employs an open-ended Q&A process. This means that any information provided by the user is accepted, and the large language model automatically determines whether the chat contains valuable information that could be included in a resume. If so, the large language model processes the collected information in the background and rewrites it into a standardized format. The rewritten content is stored in a standardized database for subsequent access and improvement. Simultaneously, the large language model summarizes the extracted information and provides feedback to the user. In an encouraging and friendly manner, the user is informed of any missing information and which sections are of higher priority. If no information is available, the large language model will point out any issues with the user's information and advise the user of the most important sections, thereby guiding the user to provide the correct response. This guided entry and conversational approach allows users to more easily complete the entire information filling process, enhancing the user experience while preventing issues like missing information and incorrect categorization. Furthermore, voice input is supported within the same chat interface. This communication method expands the information collection process to a wider range of scenarios, meeting the needs of different users in different situations.

[0031] Through the above three information collection methods, the system can obtain a user's personal information in JSON format, including basic information, education information, project experience, etc., as well as the position the user wants to apply for.

[0032] The data processing module specifically includes: 2.1 Auxiliary Information Generation: Based on the position the user is applying for, the system uses a large language model to generate two texts: job information and resume writing instructions. These two texts will serve as an external knowledge base for the large language model in subsequent resume generation, allowing the user to generate resume content that is more tailored to the position they are applying for. Job information includes the job description, job requirements, job skill keywords, and job content keywords. Resume writing instructions primarily clarify the key content to focus on in the resume, based on the job content keywords in the job information. For example, if a user is applying for a "software engineer" position, the job description generated by the large language model might include "software development" and "system design." Job requirements might include "Editing skills: Proficiency in one or more programming languages, such as Java, C++, and Python. Algorithms and data structures: Solid knowledge of algorithms and data structures, and the ability to optimize and improve algorithms." Job skill keywords might include "data structure," "database," "operating system," "Java," and "Python." Job content keywords might include "design planning," "coding and testing," "debugging and optimization," and "collaborative delivery." Resume writing instructions may include "Design Planning: Describe your experience in software design and planning, including system architecture design, requirements analysis, technical solution specification, etc. Explain how you ensure the smooth progress and high-quality delivery of projects through detailed design planning. List your specific projects and success cases in design planning." and other content.

[0033] 2.2 Text generation and optimization: The system expands and optimizes the text in the collected user information in JSON format based on the user's applied position, job information, and resume writing guidance. The collected user information in JSON format includes 8 modules: "Basic Information", "Educational Experience", "Career Experience", "Project Experience", "Scientific Research Papers and Intellectual Property", "Awards and Certificates", "Skills", and "Language". For objective descriptive information in the user information that does not need to be changed, including "Basic Information", "Educational Experience", "Awards and Certificates", and "Language", the original text in the user information remains unchanged. For subjective descriptive language in the user information that needs to be expanded and optimized, including "Career Experience", "Project Experience", "Scientific Research Papers and Intellectual Property", and "Skills", LLM is used to optimize and expand the original text based on the position information and resume writing guidance, so that the generated text is more in line with the requirements of the job application.

[0034] After data processing, the system will update the content in the obtained user information JSON to obtain the optimized user information JSON.

[0035] The resume generation module specifically includes: 3.1 Format conversion: Based on the optimized user information JSON, the system will use a large language model to convert the user information in JSON format into user information in Markdown format for subsequent resume generation.

[0036] 3.2 Template Selection: The system stores several Markdown-formatted blank resume templates and their corresponding CSS configuration files in the database. Once a user selects a resume template, the system will retrieve the corresponding Markdown blank resume template and its corresponding CSS configuration file from the database for subsequent formatted resume generation and PDF rendering.

[0037] 3.3 Resume Generation: After completing the format conversion and template selection, the system will use a large language model to fill the user's information into the corresponding positions in the blank resume template, generating a formatted initial user resume. The initial user resume format is consistent with the resume template, written in formatted Markdown syntax, and the content comes from the user's information.

[0038] 3.4 Format Check: Due to the randomness of the results generated by the large language model, the initial version of the user resume generated by the system using the large language model may have formatting issues, resulting in poor final product quality. Therefore, after the initial version of the resume is generated, the system will use the large language model again to check the format of the initial version of the user resume based on the resume template, correct any formatting issues in the initial version of the resume, and ensure that it is consistent with the format of the resume template. After completing the format check and correction, the large language will generate the final version of the user resume. This resume is written in Markdown syntax and comes with a corresponding CSS file, which can be directly rendered to generate a resume in PDF format.

[0039] The specific application process of the system disclosed in this solution is as follows: After a user enters the system, they first receive their user information JSON through the information collection module. This module includes three information collection methods: PDF parsing, AI dialogue, and page completion. After receiving the user information JSON, the system transmits it to the data processing module, which optimizes and generates the text in the user information to better meet the requirements of the position the user is applying for. In the data processing module, the system first generates auxiliary information based on the position the user is applying for, including job information and resume writing instructions. Based on this auxiliary information, the system uses a large language model to optimize the subjective descriptive text in the user information JSON, generating an optimized user information JSON. This optimized user information JSON is then transmitted to the resume generation module, which uses the optimized user information JSON to generate a Markdown-formatted user resume based on a resume template selected from a database. In the resume generation module, the system first uses the large language model to convert the JSON-formatted user information into Markdown-formatted user information. It then generates a preliminary Markdown resume based on the resume template. Finally, this preliminary resume undergoes formatting checks to produce the final Markdown resume.

[0040] The system effectively solves the following technical problems: Inefficient information collection: Through three different methods of automated information collection and analysis, the problem of traditional manual resume filling is time-consuming and cannot utilize users' existing materials is solved. It achieves the effect of efficient information acquisition, saves filling time, and solves the pain point of users taking too long to manually fill out resumes.

[0041] Resume writing is cumbersome: Leveraging the powerful knowledge retrieval and generation capabilities of large language models, resume generation and optimization are automatically performed, solving the problem of users having difficulty in quickly creating professional resumes.

[0042] Template style is monotonous: Resumes generated by large language models often have the problem of monotonous style. This system separates content generation from template selection in resume generation and uses externally stored resume templates to enhance the formatted resume generation capabilities of large language models.

[0043] Formatting errors: Large language models may make mistakes when generating complex formatted content. This system sets up an additional large language model specifically for format checking to review and correct the format of the generated resume, effectively improving the format standardization of the generated resume.

[0044] The advantages are as follows: The system collects user information from various channels, processes and refines it, and then acquires, refines, and improves effective information. Using this information and relevant resume templates, it uses a large language model to automatically generate and check resumes, helping users quickly and efficiently create resumes that meet their requirements.

[0045] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or apparatuses.

[0046] It should be understood that this solution is not limited to the specific implementation methods described above. Devices and structures not described in detail should be understood to be implemented in a common manner in the art. Any person skilled in the art can, without departing from the scope of this solution, use the methods and technical content disclosed above to make many possible changes and modifications to this solution, or modify it into equivalent embodiments with equivalent changes, without affecting the essence of this solution. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of this solution without departing from the content of this solution are still within the scope of protection of this solution.

Claims

1. An automated resume generation system based on a large language model, characterized by: It includes information collection module, data processing module and resume generation module; among them, The information collection module includes page filling, data parsing, and AI dialogue. Users manually enter basic information in the page filling part as the source data for resume generation. The data parsing part extracts relevant content from the resume document uploaded by the user to fill in the resume information. The AI ​​dialogue part extracts relevant content entered by the user in the form of chat as the resume filling information. The data obtained from these three parts are integrated into the user information database. The data processing module first applies a large language model to generate job information and resume writing guidance text based on the user's job application. Then, based on the user's job application, job information, and resume writing guidance, the large language model is used to expand and optimize the text in the user information database, and the user information database is updated. The resume generation module converts the updated user information format, selects a resume template, applies the large language model to fill the converted user information into the corresponding position of the empty resume template, generates a preliminary version of the user resume, and checks and corrects the preliminary version of the user resume before generating the final PDF format resume.

2. The automated resume generation system based on a large language model according to claim 1, characterized in that: The page filling section provides multiple resume field input items. Users manually fill in basic information and save it as source data for resume generation.

3. The automated resume generation system based on a large language model according to claim 2, characterized in that: The data parsing part extracts relevant content from the PDF documents uploaded by users to fill in resume information.

4. The automated resume generation system based on a large language model according to claim 3, characterized in that: Data parsing uses a combination of OCR and large language model analysis to automatically parse any document uploaded by the user and extract relevant information suitable for filling out a resume. Among them, OCR is responsible for scanning the text content, and the large language model will integrate, rewrite, and classify the initially acquired text. After the classification is completed, the large language model will place the acquired information into the corresponding paragraph of the standardized database as the source data for resume generation.

5. The automated resume generation system based on a large language model according to claim 1, characterized in that: For the AI ​​dialogue part, we choose to use an instant dialogue system with a large language model to conduct conversations, guiding users to complete the various information required to fill out their resumes in a chat manner. In view of the different experiences of different users and the different order of filling in information, the large language model adopts open-ended question-answering processing.

6. The automated resume generation system based on a large language model according to claim 5, characterized in that: The large language model supports user voice input in the same chat interface.

7. The automated resume generation system based on a large language model according to claim 1, characterized in that: The user information in the user information database is in JSON format. The data processing module expands and optimizes the text in JSON format. After optimization, the optimized JSON format user information replaces the original user information.

8. The automated resume generation system based on a large language model according to claim 7, characterized in that: The resume generation module first applies the large language model to convert the optimized JSON format user information into Markdown format user information. Secondly, it selects the empty resume template and its corresponding CSS configuration file. Then, after completing the format conversion and template selection, it applies the large language model to fill the content in the user information into the corresponding position of the empty resume template, generates a formatted first version of the user resume, and checks the format of the first version of the user resume. It corrects the format problems in the first version of the user resume, generates the final version of the user resume, and finally generates a resume in PDF format.

9. The automated resume generation system based on a large language model according to claim 8, characterized in that: Before the resume generation operation, several empty resume templates in Markdown format and their corresponding CSS configuration files are pre-acquired and stored.

10. The automated resume generation system based on a large language model according to claim 8, characterized in that: The final version of the user resume is written in Markdown syntax and is equipped with a corresponding CSS file.

Citation Information

Cited By

  • Resume generation method and device

    CN121303080A