Large model-based resume template generation method, apparatus and device, and storage medium

Abstract

The invention provides a resume template generation method and device based on a large model, equipment and a storage medium, and relates to the technical field of artificial intelligence such as large model and template generation. The method comprises the following steps: obtaining requirement description information for an expected resume template input by a user; inputting the required description information and the portrait information of the user as first prompt information into a preset resume template to generate a large model, and obtaining output target template components and target template styles, the large resume template generation model determines a target template component and a target template style from a pre-maintained template component library and a pre-maintained template style library based on an actual template demand identified for the first prompt information, each template component in the template component library is obtained by disassembling a sample resume template and / or is generated through a preset large model, and each template component and each template style are attached with a label corresponding to a category to which the template component and the template style belong; and combining the target template components to obtain a to-be-rendered resume template, and performing style rendering on the to-be-rendered resume template according to the target template style to obtain a target resume template corresponding to the expected resume template. According to the scheme, the difficulty of obtaining the required resume template by the user can be reduced, and the obtaining efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of data processing, in particular to the technical field of artificial intelligence such as large models and template generation, and more particularly to a resume template generation method and device based on a large model, an electronic device, a computer readable storage medium and a computer program product. BACKGROUND

[0002] Due to individual differences and demand differences, each user wants to form his own personal resume for job hunting, which usually consumes a lot of time, and it is basically difficult to directly find a ready-made resume template that completely meets his own needs. SUMMARY

[0003] The embodiments of the present disclosure provide a resume template generation method and device based on a large model, an electronic device, a computer readable storage medium and a computer program product.

[0004] In a first aspect, the embodiments of the present disclosure provide a resume template generation method based on a large model, comprising: obtaining requirement description information of an expected resume template input by a user; inputting the requirement description information and portrait information of the user as first prompt information into a preset resume template generation large model to obtain output target template components and a target template style; wherein the resume template generation large model determines the target template components and the target template style from a pre-maintained template component library and a template style library based on actual template requirements recognized from the first prompt information, each template component in the template component library is obtained based on disassembling from a sample resume template and / or by a preset large model, and each template component and template style is attached with a label corresponding to the category; combining the target template components to obtain a to-be-rendered resume template, and performing style rendering on the to-be-rendered resume template according to the target template style to obtain a target resume template corresponding to the expected resume template.

[0005] In a second aspect, the embodiments of the present disclosure provide a resume template generation device based on a large model, comprising: a requirement description information acquisition unit configured to acquire requirement description information of a desired resume template input by a user; a resume template generation large model processing unit configured to input the requirement description information and portrait information of the user as first prompt information into a pre-set resume template generation large model to obtain output target template components and a target template style; wherein the resume template generation large model determines the target template components and the target template style from a template component library and a template style library pre-maintained based on actual template requirements recognized from the first prompt information, each template component in the template component library is obtained based on disassembly from a sample resume template and / or is obtained by a pre-set large model, and each template component and template style is attached with a label corresponding to the category; and a component combination and style rendering unit configured to combine the target template components to obtain a to-be-rendered resume template, and perform style rendering on the to-be-rendered resume template according to the target template style to obtain a target resume template corresponding to the desired resume template.

[0006] In a third aspect, the embodiments of the present disclosure provide an electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement the large model-based resume template generation method described in the first aspect when executed.

[0007] In a fourth aspect, the embodiments of the present disclosure provide a non-transitory computer-readable storage medium storing computer instructions, which are used to enable a computer to implement the large model-based resume template generation method described in the first aspect when executed.

[0008] In a fifth aspect, the embodiments of the present disclosure provide a computer program product comprising a computer program, which is used to enable a processor to implement the steps of the large model-based resume template generation method described in the first aspect when executed.

[0009] The resume template generation scheme based on a large model provided by the present disclosure first acquires requirement description information of a user on a desired resume template, and then inputs portrait information capable of reflecting the user's implied requirements and the requirement description information into a preset resume template generation large model as first prompt information, so as to determine target template components and a target template style that meet actual template requirements through label matching from a pre-maintained template component library and a template style library by means of the resume template generation large model under the condition that the actual template requirements expressed by the first prompt information are fully understood, and then obtain a target resume template that is as consistent as possible with the actual requirements of the user and even meets the requirements of the user more than the expected resume template through the manner of combining and splicing each target template component and rendering the overall style according to the target template style, thereby reducing the difficulty of obtaining the target resume template and indirectly shortening the time consumption of the user in forming a required resume file on this basis.

[0010] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0011] Other features, objects, and advantages of the present disclosure will become more apparent through a detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 is an exemplary system architecture to which the present disclosure can be applied; Figure 2 A flowchart of a large model-based resume template generation method provided by an embodiment of the present disclosure; Figure 3 A flowchart of a method for processing first prompt information by a resume template generation large model provided by an embodiment of the present disclosure; Figure 4 A flowchart of a method for determining target template components and a target template style according to actual template requirements by a resume template generation large model provided by an embodiment of the present disclosure; Figure 5 A flowchart of a method for pre-constructing a template component library provided by an embodiment of the present disclosure; Figure 6 A flowchart of a method for retention detection before delivery of content filling guidance or content filling examples provided by an embodiment of the present disclosure; Figure 7 A flowchart of a large model-based resume template generation method in an application scenario provided by an embodiment of the present disclosure; Figure 8 A structural block diagram of a large model-based resume template generation device provided by an embodiment of the present disclosure; Figure 9 A structural schematic diagram of an electronic device suitable for performing the resume template generation method based on a large model is provided for the embodiments of the present disclosure. DETAILED DESCRIPTION

[0012] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to assist in understanding, which should be considered in their context only. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description. It should be noted that the embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0013] In the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solutions comply with relevant laws and regulations and do not violate public order and good customs.

[0014] Figure 1 An exemplary system architecture 100 of the embodiments of the resume template generation method based on a large model, device, electronic device and computer readable storage medium of the present disclosure is shown.

[0015] As shown in Figure 1 The system architecture 100 can include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a communication link medium between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0016] The user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. The terminal devices 101, 102, 103 and the server 105 can be installed with various applications for realizing information communication between them, such as resume template generation applications, large model-based service providing applications, instant messaging applications, etc.

[0017] The terminal devices 101, 102, 103 and the server 105 can be hardware or software. When the terminal devices 101, 102, 103 are hardware, they can be various electronic devices with display screens, including but not limited to smartphones, tablets, laptop computers, desktop computers, and the like; when the terminal devices 101, 102, 103 are software, they can be installed in the above-mentioned electronic devices, and can be implemented as multiple software or software modules, or as a single software or software module, which is not specifically limited here. When the server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server; when the server is software, it can be implemented as multiple software or software modules, or as a single software or software module, which is not specifically limited here.

[0018] The server 105 can provide various services through various built-in applications. Taking a resume template generation application that can provide a resume template generation service as an example, the server 105 can achieve the following effects when running the resume template generation application: first, receiving the requirement description information of the desired resume template transmitted by the user through the terminal device 101, 102, 103 through the network 104; then, inputting the requirement description information and the portrait information of the user as the first prompt information into the preset resume template generation large model to obtain the output target template components and target template styles, the resume template generation large model determines the target template components and target template styles from the pre-maintained template component library and template style library based on the actual template demand identified from the first prompt information, each template component in the template component library is obtained by disassembling a sample resume template and / or generated by a preset large model, and each template component and template style is attached with a label corresponding to the category; finally, combining each target template component to obtain a to-be-rendered resume template, and rendering the to-be-rendered resume template according to the target template style to obtain a target resume template corresponding to the desired resume template.

[0019] It should be noted that the requirement description information can be obtained in real time from the terminal device 101, 102, 103 through the network 104, or can be pre-stored in the server 105 locally through various ways. Therefore, when the server 105 detects that the local has already stored these data (for example, before starting to process the remaining to-be-processed resume template generation task), it can choose to obtain these data directly from the local, and in this case, the example system architecture 100 can also not include the terminal devices 101, 102, 103 and the network 104.

[0020] Since generating a resume template required by a user on demand needs to occupy more computing resources and stronger computing capability, the large model-based resume template generation method provided in each subsequent embodiment of the present disclosure is generally executed by a server 105 with stronger computing capability and more computing resources, and accordingly, the large model-based resume template generation apparatus is generally also arranged in the server 105. However, it should also be pointed out that when the terminal devices 101, 102 and 103 also have computing capability and computing resources that meet the requirements, the terminal devices 101, 102 and 103 can also complete the above-mentioned operations by the server 105 through the resume template generation application installed thereon, and then output the same result as the server 105. Especially in the case where there are multiple terminal devices with different computing capabilities, but the resume template generation application judges that the terminal device has strong computing capability and has more remaining computing resources, the terminal device can be allowed to execute the above-mentioned operations, thereby appropriately reducing the computing pressure of the server 105, and accordingly, the large model-based resume template generation apparatus can also be arranged in the terminal devices 101, 102 and 103. In this case, the example system architecture 100 can also not include the server 105 and the network 104.

[0021] It should be understood that Figure 1 the number of terminal devices, networks and servers in may be any number according to the needs of implementation.

[0022] Please refer to Figure 2 , Figure 2 A flowchart of a large model-based resume template generation method provided by an embodiment of the present disclosure is shown in FIG. 2, wherein the flowchart 200 includes the following steps: Step 201: obtaining requirement description information of an expected resume template input by a user; This step aims to obtain the requirement description information of the resume template expected by the user input by the execution subject of the large model-based resume template generation method (for example, the server 105 shown in FIG. 1). Figure 1 The purpose of this step is to first obtain the explicit expression of the user's demand, so as to provide accurate input for subsequent interaction with the large model.

[0023] To achieve the above-mentioned purpose, first, in terms of input channel, multi-modal interaction can be adopted, for example, through a text input box to receive user direct description (such as "need modern style template suitable for delivering internet product manager position"), or through a structured form to guide the user to check the key attributes (such as industry field, template complexity level, visual style preference, etc.), the combination of the two can balance the degree of freedom and data specification; Secondly, the pre-processing link of the requirement description information can also be introduced, for example, natural language cleaning mechanism can be introduced, such as filtering irrelevant symbols, correcting spelling errors, and marking user focus points through keyword extraction (such as "technology" "one page" "emphasize project experience") as the basis for subsequent fusion with portrait information.

[0024] Further, the mining of implicit requirements can also be achieved through dynamic follow-up, for example, when the user description is too brief, multiple rounds of interactive question and answer (such as "do you need a separate portfolio display block?") can be triggered, or label options can be recommended in real time based on input content (such as prompting to select "McKinsey / Roland Berger format" style preference after the user inputs "consulting industry"). In addition, the structured mapping of input information can also convert the original description into semantic units that can be parsed by large models, for example, "look professional" can be mapped to [style: business, color: cool tone, layout: symmetry], while retaining the original text as a context reference.

[0025] Further, the user input can also be standardized and replaced by a preset industry term table (such as converting the user's colloquial "IT industry" to the label library recognized [industry: internet technology]), ensuring compatibility with large models and label libraries. The final output requirement description information should contain both the original text and the structured feature vector, forming a requirement expression result that balances flexibility and machine readability.

[0026] Step 202: input the requirement description information and the user's portrait information together as the first prompt information into the preset resume template generation large model, and obtain the output of each target template component and target template style; On the basis of step 201, the present step aims to input the first prompt information composed of the requirement description information and the user portrait information into a preset resume template generation large model by the above-mentioned execution subject, and then obtain each target template component and target template style output by the resume template generation large model. Wherein, the resume template generation large model determines the target template component and the target template style from the pre-maintained template component library and the template style library based on the actual template requirement recognized from the first prompt information, each template component in the template component library is obtained by disassembling from a sample resume template and / or generated by a preset large model, and each template component and template style is additionally labeled with a label corresponding to the category. That is, the purpose of the present step is to generate a highly personalized resume template through multi-source information fusion and intelligent decision-making based on a large model.

[0027] Wherein, the construction of the first prompt information is not simply splicing user input and portrait data, but can realize deep fusion through semantic alignment and weight allocation, for example, combining the user's explicit requirement of "emphasizing project experience" with the implicit "algorithm engineer" in the portrait to generate a composite instruction with priority marking (such as [core requirement: project module highlights algorithm complexity, secondary requirement: technology stack visualization]); and the inference process of the resume template generation large model can adopt a phased strategy, that is, first parse the actual template requirement in the first prompt information through a requirement understanding module, for example, identifying that "middle managers in the financial industry" need to strengthen "management experience time axis" and "certificate display area", and then through Retrieval-augmented Generation (RAG) linkage of the template component library and the style library. It should be noted that the portrait data of the user described in the present application is recorded and stored after the user has been previously inquired and the user's expression of consent is explicitly obtained, and necessary desensitization processing is performed.

[0028] Wherein, the construction of the template component library can be realized based on a double-path manner: one is modular disassembly of a large number of sample resumes (such as decomposing a traditional resume into "educational background", "work experience", "skill radar chart" and other atomized components), and the other is to dynamically generate scarce components through an auxiliary large model (such as generating a "model parameter tuning case" component for the emerging "AI trainer" position), and all components are labeled with multiple labels (such as [function type: skill display, applicable industry: technology, experience level: senior]). The template style library can be generated by design pattern clustering, for example, encoding "modern minimalist", "vintage typewriter", "academic conference poster" and other styles into parameterizable CSS (Cascading Style Sheets) style sets, and labeling [visual weight: high, applicable scenario: creative position] and other metadata.

[0029] Further, in the target resource matching stage, the resume template generation large model can adopt a label coupling degree algorithm. For example, for the "fresh graduate + game planning" demand, the "game DEMO (example) display card" component in the component library with the [fresh graduate friendly] [portfolio priority] label is preferentially selected, and the "pixel style" style in the style library with [high color saturation] [dynamic element support] is matched. At the same time, the combination of actual conflict although the label matches can also be excluded through the adversarial filtering mechanism (such as avoiding academic type version to match cartoon icons).

[0030] Further, the target template components and the target template styles in the final output can also be attached with corresponding confidence scores. For example, the main component (core work experience module) is forcibly locked, and the secondary component (interest column) provides optional items for subsequent user adjustment, forming a generation result that takes into account automation and flexibility.

[0031] Further, the entire process described above can also be used to optimize the label system by continuously monitoring the user's editing behavior of the generated template (such as frequently deleting the "language ability" module), and then realizing the continuous updating and optimization of the component library and the style library.

[0032] Step 203: Combine each target template component to obtain a to-be-rendered resume template, and perform style rendering on the to-be-rendered resume template according to the target template style, to obtain a target resume template corresponding to the expected resume template.

[0033] On the basis of step 202, the purpose of this step is to obtain a to-be-rendered resume template by combining or assembling each target template component by the above-mentioned execution subject, and to perform style rendering on the to-be-rendered resume template according to the target template style, and then finally obtain a target resume template corresponding to the expected resume template. The purpose of this step is to organically integrate discrete modular components and stylized design into a professional-level resume template document.

[0034] To achieve this, in the combination step of the target template component, dynamic layout can be performed based on the semantic layout engine, such as automatically assigning weight areas according to component types (placing the core "work experience" component in the upper left of the visual hot area, and the "additional skills" module in the form of a folding panel in the secondary priority area), while eliminating display conflicts between components through a conflict detection mechanism (such as when "project timeline" and "skill matrix" exist at the same time, automatically switching to a left-right column layout to avoid content overflow); secondly, the structured construction of the rendered resume template can attempt to use a hierarchical abstraction method, such as first establishing a logical document tree (with "personal information" as the root node, "education background" and "work experience" as two-level branch nodes), then injecting data placeholders (such as "company name", "work time", and other field markers), and then forming the corresponding intermediate representation; in the style rendering stage, the target template style can be converted into an operational design rule set through a style interpreter, such as when the "technology blue" theme is selected, automatically applying the corresponding color palette (main color #2A5CAA), sans-serif font family, card-style shadow effect, and other global styles, while differentiating specific components (adding progress bar animations to the "technology stack" component).

[0035] Further, the rendering engine can perform multi-dimensional adaptation checks, such as including responsive breakpoint adjustments (ensuring readability on A4 paper and mobile screens), accessibility design verification (color blindness-friendly contrast, font size threshold), and industry standard compliance (such as academic resumes following APA format reference spacing).

[0036] Further, the final generated target resume template can be output in multiple version compatible formats (such as PDF format to preserve accurate layout, and HTML format to support interactive elements) as needed, with style traceability markers (recording the source of each design decision corresponding to user requirements), facilitating subsequent iterative optimization. At the same time, each process involved in this step can also be visualized through real-time preview technology to allow users to visually confirm at key nodes of component combination and style rendering, in order to better control the final target resume template.

[0037] The resume template generation method based on a large model provided by the embodiments of the present disclosure first acquires requirement description information of a user on a resume template expected by the user, and then inputs portrait information capable of reflecting the implied demand of the user and the requirement description information into a preset resume template generation large model as first prompt information, so as to determine a target template component and a target template style that meet the actual template demand of the first prompt information in a label matching manner from a template component library and a template style library pre-maintained by the resume template generation large model under the condition that the resume template generation large model fully understands the actual template demand expressed by the first prompt information, and then obtain a target resume template that meets the actual demand of the user as much as possible or even meets the demand of the user more than the expected resume template through the manner of combining and splicing each target template component and rendering the overall style according to the target template style, thereby reducing the difficulty of obtaining the target resume template and indirectly shortening the time consumption of the user in forming a required resume file on this basis.

[0038] On the basis of the above embodiment, in order to deepen the understanding of the specific processing process of the first prompt information input by the resume template generation large model used in step 203, please also refer to Figure 3 , Figure 3 The flowchart of the method for processing the first prompt information by the resume template generation large model provided by the embodiments of the present disclosure, wherein the flowchart 300 comprises the following steps: Step 301: inputting the requirement description information and the portrait information of the user as the first prompt information into the resume template generation large model; This step actually describes the integration stage of the prompt information. The above execution subject can use feature fusion technology to perform semantic alignment on the requirement description (such as “a modern style resume that needs to highlight machine learning project experience”) of the user and the portrait information (such as the historical delivery record of the user showing a preference for a technology company and an education background of a computer master), and finally form structured first prompt information through an attention mechanism to allocate weights. Of course, other implementation manners capable of achieving similar effects can also be used, which will not be enumerated one by one here.

[0039] Step 302: controlling the resume template generation large model to determine a basic template demand according to the requirement description information in the first prompt information; On the basis of step 301, this step describes the identification link of the basic template demand. In this step, the resume template generation large model will analyze the explicit elements in the requirement description, such as extracting key fields (post type: algorithm engineer, style keyword: modern) through named entity recognition, and trying to map to a preset industry template framework (internet technology post standard structure), at this time, the generated benchmark demand may contain quantitative indicators such as “the technical stack module needs to occupy 30% of the page”.

[0040] Step 303: The resume template generation large model determines the preference information according to the portrait information in the first prompt information; This step describes the extraction of preference information. In this step, the resume template generation large model will deeply mine the implicit features in the portrait. For example, according to the user's frequently used job search platform, it can infer the target enterprise size (such as frequently visiting unicorn enterprise recruitment pages to strengthen the preference for simple and practical style of start-up companies), or by analyzing the user's uploaded historical resume, it can find the long-term adherence to the layout habit (such as always using time reverse order to arrange work experience).

[0041] Step 304: The control resume template generation large model adjusts the basic template demand using the preference information to obtain the actual template demand; Based on steps 302 and 303, this step describes the use of a demand correction algorithm to correct the basic template demand combined with the preference information in the demand adjustment phase. For example, when the basic demand requires "detailed work experience" but the user portrait shows "fresh graduate", the education background module weight can be automatically adjusted and inserted into the internship project display area to form the data structure of the actual template demand (such as including module priority order, content depth coefficient, etc. Parameters).

[0042] Step 305: The control resume template generation large model determines each target template component and target template style with the label of the actual template demand in the template component library and template style library; Among them, each target template component collectively covers all template components required by the resume template.

[0043] Based on step 304, this step describes the library resource retrieval link and implements a multi-level matching strategy, that is, it can first perform accurate label matching in the template component library (such as [module type: project display] [industry: artificial intelligence]), and start similarity calculation for components that do not completely match (such as comparing the semantic distance of "machine learning" and "deep learning" labels using NLP), while in the template style library, it can use style transfer technology to dynamically adapt the user's preferred color combination (extracted from the portrait Brand color preference) to the color palette system of the selected style.

[0044] Step 306: Obtain each target template component and target template style output by the resume template generation large model.

[0045] On the basis of step 305, the step described is the output link, and the feasibility verification can also be performed at this link to ensure that each target template component can completely cover the resume elements (such as the contact information module must be included), and incompatible components can be removed through combination conflict detection (such as selecting the timeline type and education list type work experience module at the same time). Further, the final output target template component package and style scheme can be accompanied by an adaptation score, and alternative options can be retained for user fine-tuning.

[0046] To enhance interpretability, the entire process can also record the association between each decision point and the original input through the requirement traceability link, realizing the visualized explanation content of the requirement-component-style chain.

[0047] That is, the embodiment provides a resume template generation scheme through a hierarchical decision mechanism by steps 301-306, which improves the feasibility and rigor of the scheme and ensures the accuracy of the output target template component and target template style.

[0048] On the basis of the previous embodiment, the scheme mentioned in step 305 can achieve better results, please see Figure 4 , Figure 4 The flowchart of the method for determining the target template component and the target template style according to the actual template requirement provided by the resume template generation large model of the embodiment of the present disclosure includes the following steps: Step 401: Control the resume template generation large model to determine the target label corresponding to the category to which the actual template requirement belongs; In the target label determination stage described in this step, the resume template generation large model can process the actual template requirement by using a label expansion algorithm, not only to extract explicit category labels (such as [industry: finance] [post: risk management]), but also to attempt to derive derived labels (automatically add [level: intermediate] labels based on “5 years of work experience”, supplement [language: multilingual support] labels based on “apply for multinational enterprises”), and determine core labels (such as [module priority: project experience] weight 0.8) and auxiliary labels (such as [visual elements: data charts] weight 0.3) through label weight calculation, forming a multi-dimensional label vector with a confidence score.

[0049] Step 402: Control the resume template generation large model to determine the template components in the template component library with the target label as each target template component; In the component retrieval step described in this embodiment, a hierarchical screening strategy can be implemented: for example, first, a primary precise matching is performed to lock the mandatory components (such as the “compliance project evaluation” module with the [finance] [risk management] tag), and then the candidate component pool is expanded through semantic similarity calculation (such as including the [banking business] tag component in the alternatives), a new and old compatibility mode is started for components with version iterations (such as the traditional “skill list” and the new “skill radar chart”), and options that meet the user device support capabilities are retained.

[0050] Step 403: The resume template generation large model determines the template style in the template style library that has the target tag and matches each target template component as the target template style.

[0051] On the basis of step 402, this step provides a style matching with a bidirectional verification mechanism: on the one hand, the visual compatibility of the style and the component set is verified (such as the card style needs to ensure that all components support the rendering of the rounded corner border), and on the other hand, the deep fit degree of the style and the user portrait is evaluated (such as automatically filtering low-contrast color schemes for color weak users), and for the case where there is a conflict, the style adaptive adjustment is started (when the selected style is incompatible with the “portfolio gallery” component, the style main color is retained but the picture layout method is modified).

[0052] And the whole process maintains the association between the component library and the style library through a dynamic tag mapping table (such as [creative] components preferentially matching [asymmetric layout] style), and sets a fault tolerance mechanism when the main tag matching fails (such as lacking [blockchain industry] exclusive components), automatically downgrades the matching of the upper parent tag ([financial technology] components) and generates a difference prompt for subsequent optimization. The final output target resource set will be combined and checked, and the display integrity of each component under the selected style will be verified through virtual typesetting to ensure that there is no risk of content truncation or style collapse.

[0053] Through steps 401-403, this embodiment provides a scheme of intelligent tag matching and multi-dimensional compatibility verification, so as to accurately locate the resume building elements that meet the user's needs, in order to facilitate the subsequent construction of the target resume template that meets the user's actual needs.

[0054] For a deeper understanding of how to pre-construct the template component library, please see Figure 5 , Figure 5 A flowchart of a method for pre-constructing a template component library provided by the embodiments of the present disclosure, the flow 500 includes the following steps: Step 501: pre-constructing a sample resume template according to the constituent elements to obtain each original template component; This step describes the original component disassembly phase, which can specifically use dual resolution technology based on vision and semantics. For example, at the visual level, the logical blocks in the resume document are identified by layout analysis algorithm (such as identifying the left column in the PDF file as the "personal information" partition, and the upper right table as the "work experience" module), and at the semantic level, the content is functionally labeled by combining natural language processing technology (such as marking "2015-2019 XX University" as [education background] [education: undergraduate] [school type: 985]). The disassembly process preserves the layout features (such as the relative position weight in the two-column layout) and content structure (such as the "quantitative indicator of achievement" field in the project experience) of the original component, forming a set of original template components with complete metadata.

[0055] Step 502: input the generated template component generation description as the second prompt information into the preset large model, and obtain each newly generated template component; Based on step 501, this step describes the innovation component generation link. In this link, the second prompt information can not only contain component description (such as "generate a project experience module suitable for fresh graduates"), but also inject industry trend data (such as "skill graph visualization" demand). The large model ensures that the output component not only meets the design specifications (retains mandatory fields) but also has innovation (such as generating a technical evaluation module containing "open source contribution index"), and performs implementability verification on the generated result (checks whether it supports the identification of mainstream resume parsers).

[0056] Step 503: input the original template component and the fine-tuning generation description together as the third prompt information into the preset large model, and obtain each fine-tuning template component; Based on step 501, this step describes the fine-tuning component generation link, which can specifically use the iterative enhancement strategy. For example, input the original component (such as the traditional linear work experience list) and the improvement suggestion ("add job level promotion annotation") into the large model to generate a fine-tuning component that retains the core structure but has enhanced functionality (such as embedding a job level badge in the timeline), while performing version compatibility annotation.

[0057] Step 504: based on each original template component, each newly generated template component, and each fine-tuning template component, a template component library is constructed.

[0058] On the basis of steps 501, 502 and 503, this step describes the construction link of the template component library, in which multi-level quality control can be implemented, for example, the de-duplication processing can be performed first (using fuzzy matching to eliminate 90% similar components), then the cross-component association graph is established (such as automatically associating the "academic achievements" component with the "reference literature" format style), and finally the rationality of the component combination is verified by the expert system (to ensure that there is no situation of missing "practical experience" component and redundant "work experience" component).

[0059] Further, the constructed template component library can also adopt a dynamic indexing architecture to support stereoscopic retrieval according to multi-dimensional tags (function / industry / experience level), and a hot update mechanism can be set to automatically trigger the regeneration or obsolescence warning when it is monitored that the use rate of a certain type of component is continuously below a threshold.

[0060] Through steps 501-504, the embodiment provides a scheme for generating a resume module resource pool covering all scenarios by multi-source components and intelligent fusion strategy, so as to fully meet the personalized resume template needs of users.

[0061] On the basis of any of the above embodiments, please also refer to Figure 6 , Figure 6 A flowchart of a method for retaining detection based on content filling guide or content filling example before delivery is provided for the embodiment of the present disclosure, and the flow 600 includes the following steps: Step 601: For each level of template title in the target resume template, content filling guide is generated according to actual template demand and template title, and the content filling guide is filled into the content filling area under the corresponding level of template title, to obtain a target resume template to be filled; Step 602: For each level of template title in the target resume template, content filling example is generated according to actual template demand and template title, and the content filling example is filled into the content filling area under the corresponding level of template title, to obtain a target resume template to be filled; The above steps 601 and 602 both describe the generation stage of the content filling guide in the resume template, and in the practical level, the content guide strategy can be intelligently selected according to the actual template demand: for example, when the user selects the professional mode (or judges that the user is a senior practitioner), the content filling guide scheme of step 601 can be used to provide framework guidance through structured prompts (such as "work experience module is recommended to contain: ①quantitative achievements account for more than 40%; ②use industry terms; ③highlight the promotion track"); and when a novice user is identified, the content filling example scheme provided in step 602 is used to generate specific examples (such as pre-filling "leading XX system reconstruction, improving QPS by 300% through micro-service architecture" under the "project experience" title) that meet the industry specifications.

[0062] Both solutions can adopt dynamic adaptation technology to make the guidance content strictly correspond to the template title level - the main title level (such as "work experience") provides macro writing principles, and the sub-title level (such as "XX company during the tenure") provides more targeted field-level suggestions, and all pre-filled content is presented in the form of watermarks or placeholders to distinguish from the user's real input.

[0063] Step 603: In response to the user initiating a content adjustment operation on the target resume template to be filled in, determining the adjusted resume obtained after the content adjustment operation; The content adjustment monitoring described in this step can use operation flow analysis technology to track user editing behavior in real time, for example, not only record text modifications (such as deleting the company name in the example), but also identify format adjustments (such as changing a paragraph description to a bulleted list), and automatically save version change history for backtracking.

[0064] Step 604: In response to the user initiating a delivery operation on the adjusted resume, before performing the delivery operation, detecting whether all content filling areas retain original content without modification; Step 605: In response to any content filling area retaining original content without modification, not performing the delivery operation, and returning a content adjustment notification to the user.

[0065] The delivery pre-check described in the above two steps can specifically detect unmodified areas through content fingerprint comparison technology: generate a feature code (such as the semantic hash value of the example text) for each pre-filled block, and when the block feature code is found to be completely matched with the initial state, trigger a multi-level warning mechanism: 1) for light cases (such as unmodified contact information), pop up a floating prompt box to locate the problem area; 2) for serious cases (such as all work experience retaining example text), interrupt the delivery process and generate a detailed problem report (mark all areas that need to be modified, and attach a link to the industry standard resume example).

[0066] The entire process described above can be embedded with intelligent fault tolerance design, for example, when it is detected that the user has only adjusted the expression of the example text but retained the core content (rewrote "improve 30% conversion rate" to "make conversion rate increase by three times"), it is still considered as valid modification and does not trigger the block.

[0067] Further, the above execution subject can also record user high-frequency modification points (such as 80% of users will rewrite the self-evaluation module), and accordingly continuously optimize subsequent guidance / example generation strategies, forming a data-driven template improvement strategy.

[0068] That is, through steps 601-605, the embodiment provides an intelligent guidance and compliance checking mechanism to ensure that the generated resume template is both available and ensures content quality.

[0069] On the basis of any of the above embodiments, when a user's export request for a target resume template is received, the target resume template file in the corresponding format can be exported according to the export format information carried in the export request. When the export format information includes multiple different formats, the target resume template file in each format included in the export format information can be exported respectively.

[0070] That is, in the resume template export function provided in the embodiment, the above execution subject can meet the user's multi-scene use demand through the intelligent format adaptation and batch output mechanism. For example, when a request carrying export format information is received, the system first performs format demand analysis: extracts the format type (such as PDF, DOCX, HTML, etc.) and its associated parameters (such as the DPI (Dots Per Inch, resolution) accuracy requirement of PDF, the version compatibility setting of DOCX) explicitly indicated in the request, and at the same time, the implicit preferences in the user portrait (such as the user's past 90% scene selection of PDF format) are combined to perform format priority sorting.

[0071] For multi-format export scenarios, the above execution subject can start a parallel rendering engine: for printed-level PDF generation, vector graphics technology is used to ensure the sharpness of the text, metadata conforming to the standard of the job tracking system is embedded, and layout solidification checking (to prevent element displacement when opening across platforms) is performed; for editable DOCX output, the style hierarchy structure in the template is retained (CSS styles are mapped to Word style sets), and intelligent downgrading processing is performed on complex components (such as converting dynamic skill radar charts into static tables); for interactive HTML version, dynamic effects such as floating tips are retained, mobile terminal adaptive styles are automatically generated, and a resource compression package (embedded fonts, icons, and other dependencies) is attached.

[0072] And before exporting each format, a format-specific verification operation can also be performed: PDF performs text searchability verification, DOCX checks version backtracking compatibility, and HTML tests cross-browser rendering consistency. At the same time, an inter-format association relationship can be established, when any version file is modified, the key content of other format versions is updated synchronously through difference analysis (such as contact information changes need to be synchronized to all formats), and a format selection analysis report (comparing the advantages and disadvantages of each format in the target scene) is generated after the export is completed, helping users establish long-term use preferences. The entire export process uses asynchronous task queue management, large file fragmentation processing ensures system response speed, and intelligent naming services (automatically generating file names according to "name-post-date-format type") are provided when downloading the final package, and format usage guidelines (such as "PDF is suitable for formal submission, and HTML is suitable for personal website embedding") are attached.

[0073] To deepen understanding, the present disclosure also provides a complete design scheme in conjunction with one specific application scenario, which can be seen from Figure 7 The flowchart of the resume template generation method based on the large model is shown. The scheme involves the following key parts: 1) User input side technical processing: The user inputs keywords or text content of the resume template, including job information, industry information, resume style, resume color, resume format, resume module icon, and other keywords / content.

[0074] The user uploaded content first needs to be subjected to library big model risk control technology, content compliance content filtering, to ensure that there is no legal risk problem in the user uploaded content. After completing the risk control compliance audit, the library's big model text analysis technology is performed to extract the requirements for the resume template in the content. On the text side, it is directly compatible with long text and short tag input; 2) Resume template generation technology processing: After the content is refined by the input side big model analysis, it is transmitted to the library's resume template generation big model to generate a batch of resume templates that meet the requirements.

[0075] From the composition of the resume template, it includes the format, color style, module layout, and icon adaptation of the resume template. From the content of the resume template, it includes job title, school experience, work experience, personal evaluation, and skill certificate.

[0076] The number of generated resume templates: when receiving user demand, up to 4 resume templates will be generated for real-time rendering, allowing users to see the progress of the resume template generation in real time.

[0077] The generated format of the resume template: when receiving user demand, the big model will generate the resume template in HTML format, but when the user downloads it, the resume template will be transcoded to support DOC\DOCX, PDF, and PPT formats.

[0078] In actual use, users can recall AI resume template cards in the AI assistant of the application program, or directly use AI resume template generation in the AI resume template generator page. For example, AI resume template generation on a personal terminal can have the following entries: 1) Three-level classification entry of AI assistant: three-level classification of AI tool assistant page of PC library; 2) 1.2 AI assistant dialogue recall card entry: AI resume template generation card, users can directly recall AI resume template generation function in AI dialogue, which displays the AI resume template generation card and enters the AI resume template generator channel page after clicking.

[0079] By applying the scheme provided in the embodiment, the user can directly input the required template appeal of himself, and then the AI large model is used to analyze the corresponding template key points. The key points analyzed are separately processed by the AI resume template generation large model, and the required resume template and resume content are directly generated in batches. The entire process takes an average of 1-3 minutes to generate 4 complete resume templates.

[0080] Further reference Figure 8 , as an implementation of the method shown in the above figures, the present disclosure provides an embodiment of a resume template generation device based on a large model. The device embodiment corresponds to the method embodiment shown in Figure 2 , and the device can be applied in various electronic devices.

[0081] As shown in Figure 8 , the resume template generation device based on a large model 800 of the embodiment can include: requirement description information acquisition unit 801, resume template generation large model processing unit 802, component combination and style rendering unit 803. Wherein, the requirement description information acquisition unit 801 is configured to acquire the requirement description information of the expected resume template input by the user; the resume template generation large model processing unit 802 is configured to input the requirement description information and the portrait information of the user as the first prompt information into the pre-set resume template generation large model, and obtain the output of each target template component and target template style; wherein, the resume template generation large model determines the target template component and target template style from the pre-maintained template component library and template style library based on the actual template demand identified from the first prompt information. Each template component in the template component library is obtained based on the sample resume template and / or generated by the pre-set large model. Each template component and template style is attached with a label corresponding to the category; the component combination and style rendering unit 803 is configured to combine each target template component to obtain a resume template to be rendered, and perform style rendering on the resume template to be rendered according to the target template style, to obtain a target resume template corresponding to the expected resume template.

[0082] In the embodiment, in the resume template generation device based on a large model 800: the specific processing of the requirement description information acquisition unit 801, the resume template generation large model processing unit 802, and the component combination and style rendering unit 803 and the technical effects brought by them can be respectively referred to Figure 2 The related description of steps 201-203 in the corresponding embodiment will not be repeated here.

[0083] In some other optional implementations of the embodiment, the resume template generation large model processing unit 802 can include: The input subunit is configured to input the requirement description information and the portrait information of the user together as first prompt information into the resume template generation large model; The basic template requirement determination subunit is configured to control the resume template generation large model to determine a basic template requirement according to the requirement description information in the first prompt information; The preference information determination subunit is configured to control the resume template generation large model to determine preference information according to the portrait information in the first prompt information; The requirement adjustment subunit is configured to control the resume template generation large model to adjust the basic template requirement by using the preference information to obtain an actual template requirement; The component and style determination subunit is configured to control the resume template generation large model to determine, in a template component library and a template style library, each target template component and target template style having a target label to which the actual template requirement belongs; The output subunit is configured to obtain each target template component and target template style output by the resume template generation large model.

[0084] In some other optional implementations of the present embodiment, the component and style determination subunit can be further configured to: control the resume template generation large model to determine a target label corresponding to a category to which the actual template requirement belongs; control the resume template generation large model to determine, as each target template component, a template component having the target label in the template component library; wherein each target template component collectively covers all template components required to constitute a resume template; control the resume template generation large model to determine, as the target template style, a template style having the target label and matching each target template component in the template style library.

[0085] In some other optional implementations of the present embodiment, the system further includes a template component library creation unit configured to create the template component library, and the template component library creation unit can be further configured to: previously disassemble sample resume templates according to constituent elements to obtain each original template component; input the obtained template component generation description into a preset large model as second prompt information to obtain each newly generated template component; input the original template component and the fine-tuning generation description together into the preset large model as third prompt information to obtain each fine-tuning template component; construct the template component library based on each original template component, each newly generated template component, and each fine-tuning template component.

[0086] In some other optional implementations of the present embodiment, the resume template generation device 800 based on the large model can further include: The content filling guide generation and filling unit is configured to generate content filling guides for each template title in the target resume template according to actual template requirements and template titles, and fill the content filling guides into the content filling areas under the corresponding level template titles to obtain the target resume template to be filled.

[0087] In some other optional implementations of the present embodiment, the large model-based resume template generation apparatus 800 can further include: The content filling example generation and filling unit is configured to generate content filling examples for each template title in the target resume template according to actual template requirements and template titles, and fill the content filling examples into the content filling areas under the corresponding level template titles to obtain the target resume template to be filled.

[0088] In some other optional implementations of the present embodiment, the large model-based resume template generation apparatus 800 can further include: The content adjustment unit is configured to determine the adjusted resume obtained after the content adjustment operation in response to the user initiating the content adjustment operation on the target resume template to be filled. The original content residual detection unit is configured to detect whether all content filling areas retain original content without modification before performing the delivery operation in response to the user initiating the delivery operation on the adjusted resume. The content adjustment notification returning unit is configured to not perform the delivery operation and return a content adjustment notification to the user in response to any content filling area retaining original content without modification, wherein the content adjustment notification contains position information of the target content filling area retaining original content without modification.

[0089] In some other optional implementations of the present embodiment, the large model-based resume template generation apparatus 800 can further include: The export format information determination unit is configured to determine the export format information carried in the export request according to the received export request of the target resume template by the user. The format export unit is configured to export the file of the target resume template in the corresponding format according to the export format information.

[0090] In some other optional implementations of the present embodiment, the format export unit is further configured to: In response to the export format information containing multiple different formats, export the file of the target resume template in each format contained in the export format information respectively.

[0091] Corresponding to the method embodiment, the embodiment provides a resume template generation device based on a large model. First, requirement description information of a user about a desired resume template is acquired. Then, portrait information reflecting the user's implied requirements and the requirement description information are jointly formed into first prompt information, which is input into a preset resume template generation large model. By means of the resume template generation large model, target template components and a target template style that meet actual template requirements expressed by the first prompt information are determined from a template component library and a template style library in a label matching manner, so that a target resume template that meets the actual requirements of the user as much as possible, or even meets the requirements of the user more than the desired resume template, is obtained by combining and splicing the target template components and rendering the overall style according to the target template style. The difficulty of obtaining the target resume template is reduced, and the time consumption of the user in forming a required resume file on this basis is indirectly reduced.

[0092] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement the large model-based resume template generation method described in any of the above embodiments when executed.

[0093] According to an embodiment of the present disclosure, the present disclosure further provides a readable storage medium, which stores computer instructions for enabling a computer to implement the large model-based resume template generation method described in any of the above embodiments when executed.

[0094] According to an embodiment of the present disclosure, the present disclosure further provides a computer program product, which can implement the large model-based resume template generation method described in any of the above embodiments when executed by a processor.

[0095] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the present disclosure described and / or claimed in this document.

[0096] As Figure 6As shown, the device 600 includes a computing unit 601 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from the storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0097] A plurality of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, and the like; an output unit 607, such as various types of displays, speakers, and the like; a storage unit 608, such as a magnetic disk, an optical disk, and the like; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0098] The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, and the like. The computing unit 601 performs various methods and processes described above, such as the large model-based resume template generation method. For example, in some embodiments, the large model-based resume template generation method can be implemented as a computer software program that is tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the large model-based resume template generation method described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the large model-based resume template generation method by any other appropriate means, such as by means of firmware.

[0099] The various embodiments of the systems and techniques described above can be implemented in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a load programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0100] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or the block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, or entirely on a remote machine or server.

[0101] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical conductors, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0102] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0103] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0104] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. A server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of large management difficulty and weak business scalability in traditional physical host and virtual private server (VPS, Virtual Private Server) services.

[0105] According to the technical scheme of the embodiment of the present disclosure, first, the requirement description information of the user on the expected resume template is obtained, then the portrait information reflecting the user's implied needs and the requirement description information are combined to form the first prompt information, and the first prompt information is input into the preset resume template generation large model, so as to determine the target template component and the target template style that meet the actual template demand of the first prompt information from the pre-maintained template component library and the template style library in a label matching manner, and then obtain the target resume template that meets the actual needs of the user as much as possible, or even meets the needs of the user more than the expected resume template, thereby reducing the difficulty of obtaining the target resume template, and indirectly shortening the time consumption of the user in forming the required resume file on this basis.

[0106] It should be understood that the various forms of the flow shown above can be used to reorder, add or delete steps. For example, each step described in the present disclosure can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical scheme of the present disclosure can be achieved, which is not limited herein.

[0107] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A large model-based resume template generation method, comprising: obtaining user input requirement description information of an expected resume template; inputting the requirement description information and the user's portrait information as first prompt information into a preset resume template generation large model to obtain output target template components and target template styles; wherein the resume template generation large model determines the target template components and the target template styles from a pre-maintained template component library and a template style library based on actual template requirements identified from the first prompt information, each template component in the template component library is obtained based on sample resume template disassembly and / or by a preset large model, and each template component and the template style are attached with a label corresponding to the category; combining each target template component to obtain a to-be-rendered resume template, and performing style rendering on the to-be-rendered resume template according to the target template style to obtain a target resume template corresponding to the expected resume template.

2. The method of claim 1, wherein, The inputting the requirement description information and the user's portrait information as first prompt information into a preset resume template generation large model to obtain output target template components and target template styles comprises: inputting the requirement description information and the user's portrait information as first prompt information into the resume template generation large model; controlling the resume template generation large model to determine a basic template requirement according to the requirement description information in the first prompt information; controlling the resume template generation large model to determine preference information according to the portrait information in the first prompt information; controlling the resume template generation large model to adjust the basic template requirement using the preference information to obtain the actual template requirement; controlling the resume template generation large model to determine each target template component and target template style having a label belonging to the actual template requirement in the template component library and the template style library; obtaining each target template component and target template style output by the resume template generation large model.

3. The method of claim 2, wherein, The controlling the resume template generation large model to determine each target template component and target template style having a label belonging to the actual template requirement in the template component library and the template style library comprises: controlling the resume template generation large model to determine a target label corresponding to the category to which the actual template requirement belongs; controlling the resume template generation large model to determine a template component having the target label in the template component library as each target template component; wherein each target template component collectively covers all template components required to constitute a resume template; controlling the resume template generation large model to determine a template style having the target label and matching each target template component in the template style library as the target template style.

4. The method of claim 2, wherein, The creation process of the template component library comprises: pre-disassembling the sample resume template according to constituent elements to obtain each original template component; inputting the obtained template component generation description as second prompt information into the preset large model to obtain each newly generated template component; inputting the original template components and the fine-tuning generation description as third prompt information into the preset large model to obtain each fine-tuned template component; based on each of the original template components, each of the newly generated template components, and each of the fine-tuned template components, a template component library is constructed.

5. The method of claim 1, further comprising: for each level of template title in the target resume template, filling in content writing instructions according to the actual template requirements and the template title generation content, and filling in the content writing instructions to the content writing area under the corresponding level of template title to obtain a to-be-filled target resume template.

6. The method of claim 1, further comprising: for each level of template title in the target resume template, filling in content writing instructions according to the actual template requirements and the template title generation content, and filling in the content writing instructions to the content writing area under the corresponding level of template title to obtain a to-be-filled target resume template.

7. The method of claim 5 or 6, further comprising: in response to the user initiating a content adjustment operation on the to-be-filled target resume template, determining an adjusted resume obtained after the content adjustment operation; in response to the user initiating a delivery operation on the adjusted resume, before performing the delivery operation, detecting whether all content writing areas retain original content without modification; in response to any of the content writing areas retaining original content without modification, not performing the delivery operation, and returning a content adjustment notification to the user; wherein the content adjustment notification contains location information of the target content writing area retaining original content without modification.

8. The method of any one of claims 1-6, further comprising: according to the received export request of the user for the target resume template, determining export format information carried in the export request; according to the export format information, exporting a file of the target resume template belonging to the corresponding format.

9. The method of claim 8, wherein, According to the export format information, exporting a file of the target resume template belonging to the corresponding format, comprises: in response to the export format information containing multiple different formats, according to each format contained in the export format information, respectively exporting a file of the target resume template of the corresponding format.

10. A resume template generation device based on a large model, comprising: a requirement description information acquisition unit configured to acquire requirement description information of a desired resume template input by a user; a resume template generation large model processing unit configured to input the requirement description information and portrait information of the user as first prompt information into a preset resume template generation large model to obtain output target template components and target template styles; wherein the resume template generation large model determines the target template components and the target template styles from a pre-maintained template component library and a template style library based on actual template requirements identified from the first prompt information, each template component in the template component library is obtained by disassembling a sample resume template and / or generated by a preset large model, and each template component and the template style are attached with a label corresponding to the category. The component combination and style rendering unit is configured to combine the target template components to obtain a to-be-rendered resume template, and perform style rendering on the to-be-rendered resume template according to the target template style, to obtain a target resume template corresponding to the expected resume template.

11. An electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the large model based resume template generation method in any one of claims 1-9.

12. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the large model based resume template generation method in any one of claims 1-9.

13. A computer program product comprising a computer program, wherein the computer program is executed by a processor to implement the steps of the large model based resume template generation method according to any one of claims 1-9.

Images