52 results about "Markdown" patented technology

The invention discloses a plant protection field literature information batch structured extraction method based on an AI large model, and belongs to the technical field of computer data processing and artificial intelligence application. According to the method, a PDF format literature is converted into a Markdown format text through a PDF literature self-adaptive preprocessing module and a Markdown conversion module; then, a double-strategy self-adaptive AI extraction module is adopted, different modes are adopted for processing according to a text length threshold value, and JSON data are extracted in combination with a JSON format data restoration strategy; then, the CPU intensive conversion task and the I / O intensive AI analysis task are processed in parallel through a two-stage parallel scheduling module, and a structured database file is generated through aggregation according to a predefined mapping rule through a multi-dimensional aggregation export module. According to the method, end-to-end automation from literature acquisition to structured data output is realized, and the problem of low extraction efficiency of manual literature reading information is solved.

The invention belongs to the field of industry, and particularly relates to a large language model multi-modal retrieval enhanced question and answer method based on an equipment operation and maintenance document. The method comprises the following steps of: 1, converting a PDF document of equipment operation and maintenance into a document in a Markdown format; 2, generating multi-modal image subtitles on the basis of the markdown format document in the step 1; 3, firstly performing knowledge analysis based on the multi-modal image subtitles generated in the step 2, and then performing vectorization coding and similarity retrieval; and 4, by utilizing the multi-modal image subtitles generated in the step 2 and through the vectorization coding and similarity retrieval in the step 3, assisting the generation of text questions and answers based on the large language model, and realizing the large language model multi-modal retrieval enhanced questions and answers based on the equipment operation and maintenance document. The method is used for solving the problem of information analysis and understanding of various modalities aiming at equipment operation and maintenance documents which are high in speciality.

The invention relates to the technical field of automobile software testing, and discloses a software test case automatic generation method and system based on a language AI large model, and the method comprises the steps: extracting a position path, a function name and a test design direction content from user input information, generating a configuration file through a preset Prompt template, and calling the AI large model to interpret the user intention. Function, performance and regulation analysis tasks are split, an AI large model is utilized to integrate information, test design analysis is carried out, classification and redundant content cutting are carried out, self-check evaluation is completed, a formatted Markdown test design file is generated and then converted into a mind map, an engineer checks the mind map, manual modification and optimization are carried out according to needs, and the method is simple and convenient. The modified mind map is converted into a Markdown file, a formatted test case file is generated, the difference of the Markdown file before and after modification is compared through an AI large model, a Prompt template is updated, according to the method, the test case can be efficiently generated, manpower dependence is reduced, and therefore the adaptability of the test case is guaranteed.

The invention discloses a hybrid PDF (Portable Document Format) document analysis and knowledge fragment construction method and device, and belongs to the technical field of digital document intelligent processing. The method comprises the following steps: firstly, carrying out layout element identification and semantic classification on a PDF page through a deep learning model, and calculating the confidence of the PDF page; when the confidence coefficient is lower than or equal to a threshold value, a correction module based on vertical and horizontal projection analysis is automatically switched to carry out boundary detection and result error correction; the fused layout result is converted into structured JSON data, and the structured JSON data is further mapped into a Markdown format with a reserved title level; and finally, knowledge fragment segmentation is carried out based on a Markdown hierarchical relationship, so that each fragment carries a complete title context path. According to the method, the problems of semantic deficiency, insufficient stability and context splitting in the prior art are effectively solved, and the robustness and accuracy of complex format PDF and low-quality scanning copy analysis and the interpretability of downstream knowledge retrieval are remarkably improved.

The invention relates to the technical field of research and development management, and discloses a front-end code generation method and device based on a coding tool, equipment and a medium, and the method comprises the steps: extracting design manuscript element features from an original design manuscript, and constructing a graphical topological structure according to each design manuscript element feature; the graphical topological structure is converted into a structured description text through Markdown; and generating a target front-end code through the large language model and the structured description text. Through the above mode, the spatial structure understanding ability of the large language model is enhanced by extracting the design draft element features, constructing the graphical topological structure, capturing the spatial hierarchical relationship among the design elements and taking the graphical topological structure as intermediate representation, and the problem of layout deviation is solved; and the accuracy of transmitting the design intention to the large language model is improved. The method can be applied to the business fields of financial science and technology, medical health, old-age care and the like, and the reliability of generating the front-end code by the intelligent coding tool is improved.

The invention provides a traceable question-answering system for literatures in the material field and a construction method, and relates to the technical field of artificial intelligence. The method comprises the following steps: converting a PDF document into a structured Markdown text; according to the Markdown text, executing a double-stage blocking operation to obtain a blocking operation result; performing a domain adaptive vector coding operation, constructing a domain training data set corresponding to material science to train an Embedding model, and converting a text block into a vector representation according to the Embedding model in combination with a partitioning operation result; storing the vector representation in a database, and establishing an approximate nearest neighbor index structure; obtaining user question information and executing a multi-level retrieval optimization operation in the database to generate a high-confidence fragment; and splicing the high-confidence segment with user question information, and driving a field large language model to generate an answer with a precise reference mark. The problems of insufficient question-answering accuracy and poor traceability in a traceable question-answering system are solved.

The invention belongs to the technical field of data processing, and relates to a page loading method and device, computer equipment and a storage medium, and the method comprises the steps: receiving a Mar kdown file which is uploaded by a user and is matched with a target business demand; converting the Mar kdown file based on the target format to obtain target data; if the target data passes the compliance verification, performing component binding on the target data based on configuration information input by a user to obtain a configuration file; analyzing the configuration file based on a rendering engine to generate an initial page, and splitting the initial page to obtain a plurality of independent rendering units; generating priority information of each independent rendering unit; and carrying out asynchronous loading processing on all the independent rendering units based on the priority information and an on-demand loading strategy. In addition, the invention also relates to a block chain technology, and the configuration file can be stored in the block chain. The page loading method and device can be applied to page loading scenes related to insurance services in the field of financial science and technology, and the page loading efficiency is effectively improved.

Systems and methods for dynamically and automatically updating item prices on e-commerce platform are disclosed. In some embodiments, an item is offered for purchase with a current price on a website. When price elasticity data and predicted demand data for the item are both available, a first markdown price is generated for the item using a first model based on: the price elasticity data, the predicted demand data, and the current price. When the price elasticity data and the predicted demand data are not both available, a second markdown price is generated for the item using a second model based on: a decay rate, the current price, and availability of the predicted demand data. A bounded price is generated by applying an upper bound and a lower bound to either the first markdown price or the second markdown price; and transmitted to a computing device for updating the current price of the item on the website.

The invention discloses a life cycle process list calculation method based on a large language model, and the method comprises the following steps: S1, converting a literature and an image file into a. Markdown format file by using a unstructured.io visual recognition toolkit; s2, a large language model guided by cue words is used, and a table in the converted. Markdown file is decoupled; s3, reading a. Markdown file, and preprocessing the. Markdown file into a character string fragment; s4, vectorizing the character string by using an embedded model, and storing the character string by using a vector retrieval library toolkit; s5, retrieving a vector database according to the stage keyword; s6, generating a list framework by using a customized large language model according to the recalled corpus; s7, according to the input product keyword, a vector database is retrieved for the second time; s8, generating a complete list by using a customized large language model according to the recalled corpus; and S9, performing numerical value processing on the complete list in the. Json format by using a Python script, and converting the complete list into an HTML (Hypertext Markup Language) table.

The embodiment of the invention provides an original Markdown content analysis and high-speed rendering method, which belongs to the technical field of data processing, and specifically comprises the following steps: loading Markdown original content from a preset content source according to a calling scene, and uniformly processing the Markdown original content into a standard character string stream; analyzing the standard character string flow to generate an abstract syntax tree matched with a native rendering system of the swan mongolian system; mapping the abstract syntax tree into a corresponding ArkUI native component according to the type of each node in the abstract syntax tree to form a component tree; loading resources associated with Markdown original content on the basis of a native resource access mechanism of the swan Mongolia system; and rendering the component tree by using an ArkUI rendering engine of the swan gap system, and outputting the component tree to a display interface. By means of the scheme, it is ensured that the native rendering capacity with high efficiency, high consistency and high expansibility is achieved in various application scenes.

The invention discloses a method for automatically generating a test case from a product demand document by a knowledge base, which is characterized by comprising a PRD document analysis and function point extraction module, a knowledge base construction and calling module, a test case generation module and a test case generalization module, the PRD document analysis and function point extraction module is used for converting a PRD document into a structured Markdown format by utilizing a large language model; the knowledge base construction and calling module constructs a structured knowledge base by analyzing historical PRD and test case data; the test case generation module generates test cases covering a normal scene, a boundary scene, an abnormal scene and a concurrent scene; and the test case generalization module is used for generating diversified and non-repeated test cases. Compared with the prior art, the method has the advantages that an efficient and comprehensive solution is provided for software testing automation, the method is particularly suitable for function testing, boundary testing, abnormal testing and concurrent testing in the software development process, and the application prospect is good.

According to the PDF document structuring processing and conversion tool, contents (including texts, tables, images, structures and the like) in PDF documents are analyzed and converted into structured data (Markdown) capable of being used for downstream LLM tasks, and the PDF structuring processing and conversion tool is particularly suitable for PDFs of complex typesetting of academic literatures, reports and the like. Through a literature vectorization function, medical professional literatures can be precisely processed, core ideas, key data and logic relations in the literatures can be deeply understood, vectors capable of accurately reflecting the essence of the literatures can be generated, important information is prevented from being lost or misread, and information processing and application effects are improved. Meanwhile, the personalized Agent can perform adaptive optimization aiming at diversified tasks in the medical academic field, and can fit research directions, knowledge backgrounds and working habits of different users, for example, the personalized Agent provides support conforming to a rigorous logic structure and standard academic expression during academic paper writing; and when the academic slide is manufactured, concise and clear contents with highlighted key points can be presented, so that the working efficiency of medical workers and researchers is effectively improved, and the high-standard requirements of the medical field are met.

The invention discloses an automatic Markdown text conversion method and device, a medium and a program product, and the method comprises the steps: firstly analyzing a Markdown text to be converted by using a python-docx engine, and generating a basic Word document which comprises a cover, a text and a directory placeholder; then the basic Word document is stored as a temporary docx file; and then loading a temporary docx file by utilizing a win32com engine, identifying and releasing a space of a directory placeholder, inserting a real directory into the space of the directory placeholder, and refreshing an association relationship between a directory title and a page number in real time to obtain a converted document. According to the method, the advantages of Python-docx and win32com are integrated, so that automatic conversion from a Markdown text to a Word document with an automatic update directory and a standard page format is realized, and the document processing efficiency and quality are effectively improved.

The application discloses a kind of university student project education simulation method and system based on multi-agent.It includes: obtaining project creation request, creating project directory and initializing session manager and version controller, session manager maintains finite state machine, the state space of finite state machine includes three stages of system initialization, collaborative construction period and game defense period;In the collaborative construction period stage, each step is executed in turn, after completion, automatically promote state and carry out persistence, after completing review modification and auxiliary writing, enter the game defense period;In the game defense period stage, three experts parallel review and conflict detection, proposal revision based on expert opinion and final report chapter generation and cross-platform file export step are executed in turn, after proposal revision and audit pass, final report adopts the chapter independent generation strategy, and the PDF format file of Markdown and cross-platform adaptation is output.

Language models may be used for software code generation, and generation of other images and text, in notebook documents, expanding collaboration opportunities with reviewers. Further, multiple different language models may be used in various segments of the notebook, enabling collaboration among the language models. In addition to the descriptive text (markdown), software code, and rich outputs, a new prompt layer is introduced that retains, in the distributed notebook file, prompts used by language models for software code generation and / or other tasks. Thus, a reviewer does not need programming expertise to alter the software code in the notebook in order to change an output. Different language models may be used within a single notebook to leverage relative strengths (i.e., one language model for code generation and another for image generation or classification). In some examples, output from one language model may be used in a prompt for another language model.

The invention discloses a Markdown Zip packaging method and a security enhancement method, and belongs to the technical field of document processing. According to the packaging method, a Markdown document and resources are packaged into a standard Zip container, an index.md main document, an assets resource directory and a manifest.xml list are contained in the standard Zip container, and document self-containing and transplantable rendering are achieved. On the basis, the security enhancement method integrates security mechanisms such as digital signature, AES-256 encryption, integrity verification and digital watermark tracking, and solves the problems that the existing Markdown document resource is easy to lose, the path is broken, no identity authentication exists and zero protection exists. According to the method, backward compatibility is kept, meanwhile, the Markdown document has enterprise-level safety capacity, and the method is suitable for technical delivery, knowledge management and high-sensitivity scenes.

The invention discloses an automatic metadata generation method based on AI + Markdown, and provides a directory semantic analysis and content optimization scheme aiming at the problems of context length limitation of multi-document processing in a complex document and inaccurate traditional RAG information extraction. By evaluating the relevance between the catalogue and the function module, extracting related content and optimizing the length of chapters, the large generative model is called to generate structured metadata, and the structured metadata is displayed on an editing interface to be adjusted by a user. According to the method, the token consumption is reduced, the system efficiency and the automation degree are improved, and the integrity and readability of service information are ensured.

The invention relates to a Markdown-based front-end rendering enhancement method and device, equipment and a medium. The Markdown-based front-end rendering enhancement method comprises the steps that a to-be-rendered text is acquired, the to-be-rendered text is analyzed based on a preset renderer and a target plug-in, whether a user-defined tag exists in the to-be-rendered text or not is determined, and the target plug-in comprises a plug-in used for conducting semantic analysis on the to-be-rendered text; when it is determined that the custom tag exists, a tag interception operation is executed, a tag type corresponding to the custom tag is determined, and the tag interception operation is used for intercepting the custom tag; the target rendering component is determined based on the label type, the custom label is rendered based on the target rendering component, and the display interface corresponding to the custom label is obtained, so that semantic analysis of the to-be-rendered text is realized, writing of complex branch logic in the custom component is avoided, and the user experience is improved. And the code complexity and the maintenance cost of the component are high.

The invention discloses a text processing method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a to-be-processed text in response to a text input operation; the effect selection operation is responded, the to-be-processed text is analyzed based on an analysis rule corresponding to the effect selection operation, a text analysis result is obtained, and the analysis rule supports Markdown grammar and custom extension Markdown grammar; rendering the text analysis result to obtain a text rendering result; and displaying the text rendering result. Therefore, due to the fact that the analysis rule can support the Markdown grammar and the custom expansion Markdown grammar, the requirements of various different grammar can be well met, the efficiency of the follow-up rendering process can be improved, and the rendering effect can be improved. Moreover, the page does not need to be manually compiled or refreshed, so that the rendering effect can be remarkably improved, and the user experience is improved.

This invention provides a traceable question-answering system and its construction method for materials science literature, relating to the field of artificial intelligence technology. The system involves converting PDF documents into structured Markdown text; performing a two-stage block-segmentation operation on the Markdown text to obtain the block-segmentation results; performing domain-adaptive vector encoding to construct a domain-specific training dataset for materials science and training an embedding model; converting text blocks into vector representations based on the embedding model and the block-segmentation results; storing the vector representations in a database and establishing an approximate nearest neighbor index structure; acquiring user question information and performing multi-level retrieval optimization operations in the database to generate high-confidence fragments; concatenating the high-confidence fragments with the user question information to drive a domain-specific large language model to generate answers with precise citation tags. This addresses the problems of insufficient question-answering accuracy and poor traceability in existing traceable question-answering systems.

The invention provides a component document page generation method and device and electronic equipment. The component document page is generated according to a Markdown file, and the Markdown file comprises a dynamic code path. The generation method comprises the steps of obtaining an entrance component; the entry component comprises a Markdown file and an initial mapping relation comprising a dynamic code path and a component example corresponding to the dynamic code path; under the condition that the entry component is obtained, calling a pre-configured compiling method; the compiling method comprises the following steps: extracting a dynamic code path included in a Markdown file; for each dynamic code path, generating a component import code containing a component name and the dynamic code path; adding the component import code into the entry component as an update entry component; according to a compiling method, the Markdown file is analyzed according to the initial mapping relation and the update entry component, and a component document page is obtained. According to the method, the corresponding component example can be displayed at the position of the dynamic code path contained in the Markdown file, and the generation efficiency of the component document page is high.

The application discloses a life cycle process list calculation method based on a large language model, which comprises the following steps: S1, using an unstructured.io visual identification tool kit, converting literature and image files into.markdown format files; S2, using a prompt word guided large language model, decoupling tables in the converted.markdown files; S3, reading the.markdown files and preprocessing into string segments; S4, using an embedded model to vectorize the string and storing by using a vector retrieval library tool kit; S5, according to stage keywords, retrieving the vector database; S6, according to the recalled corpus, using a customized large language model to generate a list framework; S7, according to the input product keywords, secondarily retrieving the vector database; S8, according to the recalled corpus, using a customized large language model to generate a complete list; S9, using a Python script to perform numerical processing on the complete list in.json format and converting into an HTML table.

The present disclosure provides a presentation document parsing method and system, wherein the presentation document parsing method comprises: obtaining recognition results of all pages of a presentation document; wherein the recognition results comprise intermediate state JSON data and Markdown format text content of all pages; obtaining a structured information extraction outline of the presentation document according to a user intention and the recognition results; determining key pages of the presentation document corresponding to the structured information extraction outline; inputting original page images, text recognition results and chart structure codes of the obtained key pages into a visual large model to output a presentation document structured parsing result conforming to the user intention. The present disclosure generates a structured information extraction outline of the presentation document according to the user intention, so that the user does not need to know the structure and content of the presentation document in advance, and the parsing path of the presentation document can be automatically planned, greatly reducing the use threshold.

The invention discloses an audio and video multi-modal analysis system and analysis method based on large model collaboration, and aims to solve the problems of insufficient multi-modal collaborative analysis, insufficient structured metadata extraction depth and the like in the prior art. The system comprises a unified media processing agent, an algorithm processing module cluster, a large model agent cluster comprising a multi-modal content analysis / text semantic analysis / knowledge extraction / structured output agent, and a structured output module. The method covers media file receiving verification, multi-modal parallel analysis, semantic fusion and knowledge extraction, and structured result output. Through agent collaboration and localized large model deployment, multi-modal deep semantic alignment is realized, the key frame extraction accuracy rate is greater than or equal to 90%, the scene detection accuracy rate is greater than or equal to 85%, the metadata extraction integrity is greater than or equal to 95%, the speed of an asynchronous parallel architecture is increased by 2-3 times compared with that of a traditional method, a standardized JSON / Markdown report is output, and the method is suitable for intelligent analysis of audio and video contents in multiple fields.

This invention discloses a Markdown Zip encapsulation method and a security enhancement method, belonging to the field of document processing technology. The encapsulation method packages Markdown documents and resources into a standard Zip container, containing an index.md main document, an assets resource directory, and a manifest.xml manifest, achieving document self-containment and portable rendering. Based on this, the security enhancement method integrates security mechanisms such as digital signatures, AES-256 encryption, integrity verification, and digital watermark tracking, solving the problems of easy loss of Markdown document resources, broken paths, lack of authentication, and zero protection in existing Markdown documents. This invention, while maintaining backward compatibility, endows Markdown documents with enterprise-level security capabilities, suitable for technology delivery, knowledge management, and highly sensitive scenarios.

This application discloses an automatic generation system and method for XSD interface parameter documents for knowledge bases, belonging to the field of data processing and storage technology. It includes: an input validation module for validating the validity of user-inputted XSD file path, Markdown output path, interface description information, and root node name; a file parsing module for parsing the XSD file and extracting interface parameter information; a parameter processing module for recursively processing the extracted interface parameter information and constructing a complete parameter path; a document generation module for generating standardized Markdown documents based on the parameter paths; a segmentation processing module for intelligently segmenting the generated Markdown document content; and an output control module for saving the segmented Markdown documents to a specified path. This solution addresses the problem of incomplete parameter path information in existing technologies, avoids semantic loss caused by segmentation processing, ensures a unified output format, and facilitates subsequent processing and integration.

The invention discloses a multi-system API collaborative intelligent scheduling method based on a large model, and the method specifically comprises the steps: recognizing a theme intention of a user request through a large language model and a mixed attention mechanism, and outputting a corresponding theme type and a business entity; the service API description is vectorized and stored in a vector database, mixed retrieval is carried out by combining a user input vector and a topic vector, and a plurality of most relevant APIs are found. For each candidate API, a request parameter is generated based on the API description and the user context, and a supplemental question process is triggered when the parameter is incomplete. After the parameters are complete, API calling is executed, and a response result is processed. Responses, service roles, user contexts and prompt words of a plurality of APIs are input into a large language model together, credibility weighting, enhancement and consistency analysis are carried out, and a unified reasoning result is generated. And the system is pushed through the server and carries out rendering output in a Markdown, HTML, JSON or graphic mode.

An object of the system according to the embodiment is to further receive the benefit of the generation AI by using the GUI-based content creation tool using the Markdown and the CSS.SOLUTION: A system according to an embodiment includes a generation AI, a GUI-based content creation tool, a Markdown inputter, and a CSS inputter. The generation AI analyzes the text inputted by the Markdown inputting unit. The GUI-based content creation tool creates a figure or slide based on the text parsed by the generation AI.SELECTED DRAWING: Figure 1

The invention provides an intelligent document analysis and enhanced retrieval method and system and a readable storage medium. The intelligent document analysis and enhanced retrieval method comprises the steps that an original document is converted into a Markdown format document through an analyzer; the method comprises the following steps: analyzing the content of a Markdown format document through a field large model, and intelligently segmenting the Markdown format document according to the continuity of a technical theme to generate segments with complete semantics; and segmenting the segmented segments through a domain model trained by the automobile text to obtain a retrieval unit. By means of the technical scheme, multi-topic mixing can be eliminated, retrieval precision and answer correlation are improved, segmentation is guided through domain knowledge, the completeness of technical terms and technical description is guaranteed, and the accuracy of generated answers is greatly improved.

Language models may be used for software code generation, and generation of other images and text, in notebook documents, expanding collaboration opportunities with reviewers. Further, multiple different language models may be used in various segments of the notebook, enabling collaboration among the language models. In addition to the descriptive text (markdown), software code, and rich outputs, a new prompt layer is introduced that retains, in the distributed notebook file, prompts used by language models for software code generation and / or other tasks. Thus, a reviewer does not need programming expertise to alter the software code in the notebook in order to change an output. Different language models may be used within a single notebook to leverage relative strengths (i.e., one language model for code generation and another for image generation or classification). In some examples, output from one language model may be used in a prompt for another language model.