56 results about "Specification document" patented technology

The invention discloses an engineering design document compliance inspection method, electronic equipment and a storage medium, and the method is based on an OCR and a large language model, and comprises the following steps: standardizing the character recognition and extraction of a document; standardizing character error check and manual correction of the text; performing structured processing on the corrected standard text; vectorization processing of core text content and construction of a standard query system; presetting a specification set range corresponding to each chapter of the design document; carrying out vectorization and similarity query on characters selected by a user in the text editor; corresponding standard content is provided for the user; the method can improve the efficiency and accuracy of specification reference in the document, facilitates the dynamic update management of specification contents, and facilitates the guarantee of the correctness of reference specifications.

The invention provides a demand specification generation method and system based on multi-modal understanding, and relates to the field of artificial intelligence, and the method comprises the steps: carrying out the content understanding operation of data of each modal through employing a corresponding content extraction tool, so as to generate a demand representation document after multi-modal unification; establishing a demand knowledge base according to an existing template base and a corresponding rule base, matching and comparing the demand knowledge base with the demand representation document, checking the integrity of the demand representation document, and generating a question for seeking a decision; creating an overall framework of the standard document according to the standard standard template, and performing content filling on the overall framework according to the demand representation document and an answer to the question seeking the decision to generate the standard document; and collecting feedback information for the standard document, analyzing the feedback information by using a large language model, and converting the feedback information into executable modification suggestions for the standard document. According to the method, the automation degree, the integrity and the accuracy of demand specification generation are improved.

A method and system process a construction domain query. A user query relating to a construction specification document is obtained. The construction specification document is obtained A Table of Content (ToC) page detection module autonomously classifies each page of the construction specification document as a table of content (ToC) or not. A text extraction module autonomously extracts text content from the construction specification document. A document sectioning module autonomously outputs section titles based on the ToC page classification results and the extracted text content, and sections the construction specification document into sections based on the section titles. The user query is processed based on the sectioned construction specification document.

The invention provides an automatic software development system based on multi-agent cooperation, and the system comprises a demand analysis agent which is used for receiving a natural language demand input by a shared working area, generating a demand specification document containing an acceptance standard based on the natural language demand, and sending the demand specification document to the shared working area; the detailed design agent is used for receiving a demand specification document input by the shared working area, performing system design based on the demand specification document, generating a design specification document and sending the design specification document to the shared working area; the project development agent is used for receiving a design specification document input by the shared working area, performing code development based on the design specification document, generating an executable code and sending the executable code to the shared working area; and the shared working area is used for storing and forwarding the natural language demand input by the user and the output of the demand analysis intelligent agent, the detailed design intelligent agent and the project development intelligent agent. The development efficiency can be improved, the labor cost is reduced, and the software quality is improved.

The invention relates to a 5G security demand extraction and assertion generation technology based on LLM and RAG, and the technology comprises the following steps: 1, a standard document preprocessing module, splitting a 3GPP standard document according to a security process hierarchy, mapping the 3GPP standard document to a corresponding label, and solving a knowledge fragmentation problem; the RAG knowledge base construction module analyzes document slices by using an RAGF low tool, divides the document slices into text blocks, stores the text blocks into a database through vectorization of an embedded model, and realizes efficient retrieval by adopting hierarchical indexing; 3, an LLM semantic understanding and assertion generation module, which is used for converting a natural language demand into a logic expression based on a DeepSeek model in combination with a cue word framework, a knowledge base and a signaling alphabet; and the assertion post-processing module is used for performing grammar check, standard conformity verification, simplification and logic conflict detection on the LTL assertion. According to the technology, automatic extraction of security requirements and assertion generation are realized, efficiency and accuracy are improved, security requirements in multiple aspects are covered, and a reliable scheme is provided for 5G terminal protocol security testing.

The invention relates to an agent generation method and system for enterprise process automation. The method comprises the following steps: S1, obtaining a standard specification document of an enterprise; s2, knowledge extraction and knowledge graph construction; s3, semantic analysis and intention discrimination; and S4, process reasoning and automatic execution. According to the method, full-link intellectualization from enterprise system document automatic analysis to business process closed-loop execution is realized, context sensing and dynamic adaptation capabilities are realized, multiple types of tasks are supported, the analysis fine granularity and cross-system cooperation efficiency of business process rules are effectively improved, and high universality and expandability are realized.

The invention provides a multi-agent collaborative supply chain bill system automatic development method and device, and the method comprises the steps: receiving a user demand document related to a to-be-developed supply chain bill system, and obtaining an enterprise specification document used for constraining development technical specifications; initializing a workflow arrangement module and a state data structure for automatically developing the supply chain bill system; wherein the workflow arrangement module comprises a directed graph, an intelligent agent is used as a node, and a circulation rule is used as an edge; and the intelligent agent processing module is used for enabling the intelligent agents to work serially and / or concurrently according to an execution path determined by the workflow arrangement module by taking the enterprise specification document as a constraint, analyzing the user demand document to obtain a design document, planning a development task according to the design document, generating review opinions by reviewing the planned development task, and sending the review opinions to the intelligent agents according to the review opinions. And generating a code file based on the design document, the development task and the review opinion, and compiling the code file to obtain an executable file related to the supply chain bill system.

The invention relates to an engineering field knowledge graph construction method and system. The engineering field knowledge graph construction method comprises the following steps: S1, obtaining an engineering specification document and a contract document; s2, extracting technical elements from the engineering specification document, and constructing a technical knowledge sub-graph which comprises mutually associated technical parameter nodes; s3, responsibility elements are extracted from the contract document, a contract knowledge sub-graph is constructed, and the contract knowledge sub-graph comprises mutually associated contract term nodes; according to the method, information islands are broken, three-layer knowledge graph fusion is carried out on the project specification document, the contract document and the project decomposition structure, three-dimensional penetration association of technical parameters-contract terms-project entities is established, automatic mapping of the technical parameters, the contract terms and the project entities is realized, and the project quality is improved. The efficiency of engineering technology compliance check and contract risk identification is greatly improved, and low efficiency and high error rate of traditional manual management are avoided.

This disclosure relates to method and system for facilitating legacy code transformation. The method includes receiving legacy code data and natural language document from one or more data sources. Each of the one or more data sources is one of an external data source or an internal data source. Further, the method includes generating a first natural language output based on the legacy code data through a first LLM, and a second natural language output based on the natural language document through a second LLM. Further, the method includes fine-tuning one of the first LLM or the second LLM based on the first natural language output and the second natural language output, through a third LLM. Further, the method includes generating a natural language specification document corresponding to the legacy code data based on the first natural language output and the second natural language output through the third LLM.

This invention discloses a method and system for generating real-time content in a specific domain based on AI technology, belonging to the field of artificial intelligence technology. The method includes the following steps: 1. Constructing a local material knowledge base RAG_src in the form of word vector storage using AI technology, whose materials include multimodal materials such as text, voice, audio / video, or H5; 2. Constructing a local domain knowledge classification knowledge base RAG_policy in the form of file storage using AI technology, whose content includes domain specification documents and technical classification documents categorized by domain knowledge and information; 3. Constructing PageIndex by combining a Large Language Model (LLM) with RAG_policy; 4. Receiving content generation requests from real-time software, calling RAG_src and PageIndex through LLM to generate real-time content that meets the requirements of the corresponding scenario; 5. Checking the generated content and outputting compliant content. This invention solves the problems of non-standard content generation, disconnection from relevant standards, low reuse rate of proprietary materials, and low retrieval efficiency in existing technologies, achieving accurate, efficient, and real-time content generation. It is adaptable to various scenarios such as real-time content generation, content compliance checking, and real-time domain assistance in specific domains. Application scenarios include, but are not limited to, oral practice and oral examinations according to teaching syllabi.

The invention discloses a platform product optimization and research and development process remodeling method based on AI drive, comprising the following steps: constructing a general MCP service: dynamically generating an OpenAPI standard document through API scanning registration; analyzing the user language and generating a visual chart or a text abstract; analyzing the natural language, and converting the structured data returned by the API into the visual chart or the text abstract for structural integration and feedback; and establishing a non-invasive AI enabling mechanism: intercepting and processing an API calling request, and providing an interaction component through front-end SDK embedding. Through cooperation of the MCP service and the WebAI assistant, low-invasion AI enabling of a traditional platform product is achieved, and the problems that API integration cost is high, task execution is fragmented and the like are solved. And in combination with a plug-in architecture, the platform can quickly expand the AI capability, and the research and development efficiency and the user experience are remarkably improved.

This invention provides an automated compiler testing method based on specification inference and loop consistency verification to address the problems of traditional compiler testing, which relies on differential testing, requires reference implementations, and cannot effectively test domain-specific compilers. The method first extracts a syntactic skeleton set and a detailed reference set from heterogeneous specification documents; it then uses a large language model to infer type-safe formal grammar rules through multi-agent collaboration; based on the inferred grammar, it generates semantically valid test cases using a seed-preserving abstract syntax tree-level mutation strategy; it executes a compile-decompile-recompile loop on the test cases and detects compiler defects through binary-centralized comparison and noise normalization. This invention achieves systematic automated testing of compilers in the absence of reference implementations and execution oracles, improving compiler reliability and system security.

The invention discloses a software engineering implementation method, and the method comprises the steps: business modeling: anchoring a demand source, and converting a business demand into a landing model; the software requirement specification specification is used for solidifying a requirement standard and forming a specification document based on the business modeling result; summary design: constructing a technical framework, joining demands and development, and macroscopically planning a system architecture; the framework is converted into an executable scheme, and coverage integration, implementation and development are achieved. According to the software engineering implementation process, a standardized'business modeling driving type 'specification is taken as a core, a five-level process model and a data model are used for constructing a business architecture core framework, full-link connection of'business demand-data logic-IT design-landing implementation' is broken through by means of a structured view, and complex processes / functions are deconstructed and split into micro-service interface-level granularity, micro-service interface-level granularity, micro-service interface-level granularity and micro-service interface-level granularity. Accurate correspondence and decoupling of the service component and the IT component are achieved, and finally efficient landing of the service architecture in the IT architecture is guaranteed.

The invention discloses an interface document generation method and device, a storage medium and a program product, and relates to the technical field of machine learning. Key metadata corresponding to the input and output tensor of the target model is adaptively extracted according to the type of the target model, and the data type of the key metadata is converted according to a mapping conversion rule of a pre-established model framework native data type and a service interface standard data type; and finally, according to the key metadata after data type conversion, generating an interface specification document of the target model. According to the method, the metadata of the to-be-deployed target model is automatically analyzed, and the native data type of the model framework is mapped and converted into the data type of the service interface standard, so that the interface specification document of the target model is further generated, the compiling output efficiency of the API document is high, errors are not easy to occur, the maintenance cost is remarkably reduced, and the development efficiency of the target model is improved. And synchronous updating of the document and the model version is ensured.

Mechanisms to transform a multimodal hardware specification document into register-transfer level (RTL) code by configuring a large language model into multiple agents to transform the hardware specification document into a structured implementation plan comprising a sequence of hardware functions, configuring the large language model to apply progressive coding and prompt optimization to sequentially transform the hardware functions into low-level program code through a plurality of progressively lower-level code generation stages, and transforming the low-level program code of the hardware functions into the RTL code through a code optimizer and high-level synthesis tool.

The embodiment of the invention relates to the technical field of document processing, and discloses a function specification document design method and device, electronic equipment and a storage medium. The method comprises the following steps: after a function model is debugged, decomposing a model structure and a logical relationship of the function model to determine a chapter structure of a function specification document; designing a cover template, and obtaining data source information of the functional model; compiling files corresponding to the cover template and the functional structure based on the data source information and the chapter structure; and compiling and generating a function specification document based on all files. According to the method and the device, the problems of low efficiency and high error rate of a mode of manually compiling the function specification document are at least solved, the compiling efficiency of the function specification document is favorably improved, and the design error risk is reduced.

The invention relates to a protocol fuzz test knowledge automatic construction method based on a large language model, and belongs to the technical field of network security testing. The method comprises the steps that a standard document of a target protocol is preprocessed, and a document fragment set with a document structure is output; constructing a basic protocol intermediate representation (IR) from the document fragment set by using a large language model; inputting the basic protocol IR and a fuzzy test task indication into a large language model together, and constructing a fuzzy test IR comprising a field semantic category, a normalized constraint description and a fuzzy test related strategy prompt; writing the fuzzy test IR into a protocol fuzzy test knowledge base, and establishing a hierarchical storage and index structure; when the fuzzy test engine executes the fuzzy test task, knowledge in the knowledge base is called to guide generation and variation of the test case set. By using the method provided by the invention, rapid adaptation and protocol perception fuzz testing of multiple protocols can be realized, and the automation degree and vulnerability discovery capability of the protocol fuzz testing are remarkably improved.

The invention provides a 5G O-RAN component automatic test method and system based on protocol perception. The method comprises the following steps: firstly, based on an O-RAN normative document protocol, extracting functional definition information, participating entities and communication interface information of a target component, and generating an expected signaling process in combination with a retrieval enhancement generation technology and a large language model; then, based on the communication test case in the expected signaling process, executing a communication test on the to-be-tested O-RAN component, and obtaining an actually executed standard communication process through network flow collection and protocol analysis; and finally, comparing and analyzing the standard communication process and an expected signaling process, and outputting test results including signaling sequence abnormity, message missing and message errors. According to the invention, automatic testing of the communication behavior of the O-RAN component is realized, manual intervention is reduced, the testing efficiency and accuracy are improved, the interoperability verification capability among multi-manufacturer components is enhanced, and the communication reliability and stability of a 5G O-RAN system are improved.

The present application is applicable to the field of artificial intelligence technology, and provides a method, apparatus, device and storage medium for determining interface specifications. It includes: collecting multiple interface specifications; inputting multiple interface specifications into a specification evaluation model for processing to obtain a development scenario corresponding to each interface specification; determining multiple groups of interface specification documents based on the development scenario corresponding to each interface specification; when a development interface request is received, determining the actual development scenario corresponding to the development interface request; and determining a target interface specification document that matches the actual development scenario in multiple groups of interface specification documents. In the above scheme, the target interface specification document for developing the interface is quickly generated and determined, without the need for the development parties to discuss the interface specification back and forth many times, simplifying the steps of developing interface agreement specifications, thereby improving development efficiency and saving development costs.

The invention discloses an industrial control protocol difference fuzzy test method and system based on protocol consistency verification. The method comprises the following steps: firstly, based on an official specification document of a target industrial control protocol, constructing a formalized baseline protocol finite-state machine, and generating a consistency verification sequence set covering protocol core logic; secondly, performing consistency testing on a plurality of tested protocol implementation entities, identifying and recording behavior deviation between each entity and the baseline model, and generating inconsistency judgment information; further, a differential fuzzy test case is constructed by utilizing the judgment information to generate an auxiliary vector, a targeted variation strategy is designed, and a final differential fuzzy test case set is generated and optimized; and finally, executing testing in parallel in an isolated virtualization environment, and automatically judging abnormal entities and vulnerability trigger points corresponding to the abnormal entities by performing vectorization processing and similarity analysis on response logs of the entities. According to the method, the problems of target fuzziness, low efficiency and high false alarm rate of an existing fuzzy test method are effectively solved, and the accuracy, the efficiency and the automation level of industrial control protocol security testing are remarkably improved.

Techniques are described for generating an API proxy and backend implementation. An example method can include accessing computing resource modeling information and data element relationship modeling information. The method can further include generating an API outline comprising a field identifier and a field definition. The method can further include generating an OAS document comprising information for defining an application programming interface proxy. The method can further include accessing API proxy information from the open application programming interface specification document. The method can further include generating an API proxy based at least in part on the API proxy information, the API proxy operable for accessing information based at least in part on the field identifier or the field definition.

The invention provides an intelligent auxiliary process document generation method and system based on a ring forging, and is applied to the technical field of data processing. Comprising the following steps: generating target product specification document information according to a product specification basic information form and a technical standard system library of a ring forging process of the ring forging piece; constructing a stock process database according to the stock process data of the ring forgings; basic data information of the ring forging in the stock process database is processed to obtain product features, material features, equipment features and target specification features, and then initial forging step parameters are determined; according to the four features and initial forging step parameters, ring forging input features are constructed and input to a defect detection model, and a ring forging classification result is obtained; and when the classification result of the ring forgings indicates that the ring forgings do not have defects, the initial forging step parameters serve as forging step parameter information, and then product process document information is generated in combination with the target product specification document information. And the compiling accuracy and efficiency of the process document information are improved.

The embodiment of the specification relates to a specification driven coding method and device, and the method comprises the following steps: in response to a change operation of a user on a specification document, determining change content in the specification document; according to the change content, updating a corresponding specification graph, and determining a plurality of changed change requirement nodes; the specification graph contains a plurality of types of nodes, and corresponds to the semantics of the specification document; according to any first change requirement node, determining a plurality of first recall nodes with a high ranking relevance in the updated specification graph; according to the first change requirement node and the plurality of first recall nodes, determining a first change context; inputting the first change context into a large language model to perform a first update on a current code library.

Apparatuses, systems, and methods for extracting structured specification requirements from specification documents associated with a device under test (DUT) can include generating one or more semantic units based on DUT specification documentation and performing a structured large language model (LLM) call to generate test requirements for the DUT based on the generated one or more semantic units and an entity extraction task. The entity extraction task can be defined via a system prompt command that is responsive to presenting an end user with a system prompt for the entity extraction task. The one or more semantic units can be generated via portioning of DUT specification documentation.

The invention discloses a code generation method and device, equipment and a storage medium. The method comprises the steps that at least one target natural language instruction matched with original natural language description input by a user is acquired from an industrial semantic library; generating a PLC code file according to the PLC code sample, the logic label and the standard document associated with the at least one target natural language instruction; and converting the PLC code file into a PLC image. PLC code samples, logic tags and standard documents associated with various different natural language instructions are stored through an industrial semantic library, so that the reliability of basic data is ensured, human errors are reduced, and therefore, basic support is provided for original natural language description input by a user by using the industrial semantic library; according to the method, a natural language is accurately converted into a compliance code based on semantic-logic mapping, the generated code is converted into a PLC image, millisecond-level downloading is achieved, automatic starting and running are achieved, a complete technical link is formed, manual programming and debugging time is saved, and code development efficiency and code quality are remarkably improved.

The invention discloses an evaluation double-layer directory construction method and system based on an electric power AI application scene. The method comprises the steps of obtaining a business specification document; the business specification document is analyzed to obtain core elements, and the core elements comprise business scenes, observation objects and attributes of the observation objects; constructing a multi-level nested structure based on the core elements to obtain a preliminary double-layer directory; reviewing and adjusting the initial double-layer catalogue to obtain an adjusted double-layer catalogue; verifying and perfecting the adjusted double-layer catalogue by using historical business data to obtain a corrected double-layer catalogue; and finally auditing the corrected double-layer catalogue to obtain an evaluated double-layer catalogue. By implementing the method provided by the invention, three major problems of fuzzy structure hierarchical relationship, lack of business and technical team collaborative semantic framework and high construction cost of the traditional knowledge graph in vertical business scene application can be solved, so that efficient and low-cost knowledge graph construction which is closely combined with business logic is realized.

Apparatuses, systems, and methods for extracting structured specification requirements from specification documents associated with a device under test (DUT) can include generating one or more semantic units based on DUT specification documentation and performing a structured large language model (LLM) call to generate test requirements for the DUT based on the generated one or more semantic units and an entity extraction task. The entity extraction task can be defined via a system prompt command that is responsive to presenting an end user with a system prompt for the entity extraction task. The one or more semantic units can be generated via portioning of DUT specification documentation.

A method or system for processing a specification document associated with a circuit design to identify design requirements. The method includes receiving a specification document associated with a circuit design, and processing the specification document to identify at least one of a text component, a table component, or a finite state machine (FSM). After that, the text component is parsed by a first parser to identify a first set of design requirements. The table component is parsed by a second parser to identify a second set of design requirements. The FSM component is parsed by a third parser to identify a third set of design requirements. The identified first, second, and / or third set of design requirements are then provided for display to a user for review.

The invention discloses a document analysis and maintenance system and method based on a large model and semantic understanding, and the system comprises the following modules: a data access module which accesses a QMS system, obtains a standard field set needing to be matched at present, and accesses a multi-modal specification document uploaded by a user; the specification standardization module is used for uniformly carrying out structured processing on the specifications; the semantic mapping maintenance module is used for storing a mapping rule of a QMS standard field and a synonym set of the QMS standard field; the specification information extraction module is used for performing structured information extraction on the specification subjected to structured processing by utilizing a large model; the algorithm verification module is used for performing semantic matching verification and specification value standardization calculation on the extraction result; and the user auditing interaction module is used for providing man-machine collaborative auditing and optimization feedback for the user. According to the invention, end-to-end automatic processing from a heterogeneous specification to QMS system structured data can be realized through cooperative work of six functional units. Each unit is clear in responsibility, and decoupling design is achieved.

The invention provides a card making data processing software development and management system based on artificial intelligence, which comprises a demand management module, a software project management module, a development auxiliary module, a test data and use case management module and a full life cycle management module, the development auxiliary module is used for receiving the key points transmitted by the software project management module and performing code generation, annotation and fragment labeling by using LLM; meanwhile, through a built-in knowledge base self-iteration mechanism, dynamically updating a code knowledge base according to a code generation and optimization process; the test data and case management module generates large-batch test data and test cases through LLM and a dynamic card making data rule base based on small-batch card making data and description documents thereof provided by customer demands, and executes an automatic test process. According to the method, the development efficiency of the card making data processing software can be effectively improved, the coverage rate and effectiveness of software testing are improved, the software maintenance and management cost is reduced, and the risk caused by pure manual development is reduced.