58 results about "Canonicalization" patented technology

This application discloses a method, system, and device for detecting malicious code variants using control flow graph semantic matching, belonging to the field of computer network security technology. It includes: disassembling the target executable file, constructing a control flow graph, and extracting basic block semantic features to generate semantic vectors; obtaining semantic equivalence classes based on semantic vector clustering, identifying main anchor points, and constructing a main anchor point skeleton graph; performing semantic compression and noise processing with the anchor point skeleton as constraints, and obtaining a semantic core graph through topological verification; hierarchically normalizing the semantic core graph and the main anchor point skeleton graph to generate a composite hash signature; and matching the composite hash signature with a malicious code family signature library to complete variant determination. The above scheme, through the main anchor point constraint compression and normalization process, resists structural obfuscation disturbances, improves the consistency of homologous variant identification, and has advantages such as strong robustness, high efficiency, and good interpretability. It is suitable for detecting malicious code variants using control flow graph semantic matching in complex obfuscated environments.

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 application discloses a multi-scale structure preserving optimal transmission depth kernel clustering method, and belongs to the technical field of unsupervised machine learning, pattern recognition and data mining, and comprises the following steps: S1, multi-layer coding is performed on input unlabeled single-view data to obtain hierarchical hidden representation; S2, each layer of normalized kernel matrix is obtained based on the step S1; S3, a fusion kernel matrix is obtained based on the step S2; S4, the fusion kernel matrix obtained based on the step S2; S5, a multi-scale structure preserving regularization term of the step S3 is taken as a core constraint; and S6, after the final fusion kernel matrix is obtained after model training convergence, a multi-scale structure preserving regularization is constructed from the local neighborhood distribution consistency and the global geometric correlation, the geometric deviation of the kernel space relative to the input space is explicitly constrained, and the excessive geometric distortion and structure degradation in the kernel learning process are inhibited.

The application provides a system and method for parsing PLC intermediate code based on compilation technology and automatically generating data bridge description. The system comprises: a preprocessing module for normalizing the C intermediate code generated by matIEC; a lexical analysis module for converting the code into a Token stream; a syntax analysis module for constructing an abstract syntax tree; an abstract syntax tree processing module for traversing the AST to extract type definitions, fields and memory layouts, and converting them into an internal data model; a code generation module for serializing the data model into a standardized JSON file, including data item name, type, size, offset, memory type and nested sub-items; a version management module for automatically updating the compiler version number and embedding the JSON based on CMake during building. The application realizes full-automatic extraction and standardized output of PLC data structure, eliminates manual parsing, ensures accurate memory layout, supports cross-platform integration and version consistency verification.

A formal system effects ordered derivations of canonical forms with inductive formation rules. A logical connective understood as ordered conjunction, denying the proof-theoretic structural law of exchange in a substructural setting, concatenates terms occurring at successive internal nodes of a proof tree until a defined pattern is matched and a canonical form is thus obtained. Leaf nodes classify subterms by their semantic types, and syntactic composition occurs at internal nodes to determine the canonical form. The framework admits normalization of structured data, deriving consistent representations for terms with many possible encodings.

The application relates to the technical field of natural language processing, in particular to an AI generated text detection method and system based on a deep learning model. The method comprises the following steps: performing character normalization and sentence segmentation on a to-be-detected text, segmenting the text by a sliding window to generate a segmented sequence; constructing deep semantic discrimination features, logarithmic probability difference trace features, style structure features and entity consistency features; inputting a gate fusion discrimination network, outputting a generated probability and confidence degree after probability calibration, and outputting a segmented contribution degree; performing threshold adaptive determination according to a risk level, a text length and the confidence degree, outputting a review state and generating a traceable record when the confidence degree is insufficient; and triggering incremental updating based on drift monitoring and a feedback sample pool. The application improves cross-domain robustness and interpretability, reduces false positives and false negatives, and supports long-term stable operation.

The present application relates to a kind of structure inference, automatic analysis and standardized output of multi-variant BOM table, belong to electronic manufacturing data processing, table analysis and structured data cleaning field.The present application constructs a set of analysis architecture for multi-variant BOM table by the modular processing link of "title row detection-column head normalization-multi-column identification-variant reconstruction-segment cleaning-intelligent inheritance-standardized output", robustly locates complex table head by strong and weak evidence weighting and verification under-probing mechanism, realizes the split and identification of multiple quantity columns using stop mark and semi-finished product number analysis, and introduces cross-row backfilling and position number inheritance rules under paragraph constraint, effectively solves the analysis problem caused by template difference, multi-column variant and information loss, realizes the automation, precision and standardization of BOM data from messy input to engineering level standardized output, significantly reduces manual intervention and improves data quality.

The application discloses an engineering material intelligent inquiry matching method and system based on natural language processing. The method comprises the following steps: obtaining an inquiry request text input by a user; performing named entity recognition by using a pre-trained language model in the field of engineering cost to extract key entities such as material name and specification model; performing semantic normalization processing through a domain ontology knowledge base; encoding the normalized conditions into vectors and combining structured fields to perform hybrid retrieval; performing adaptive statistical outlier filtering on candidate prices; calculating a confidence score according to the data volume, timeliness, semantic matching degree and data source label by using Bayesian confidence estimation, and outputting a recommended result in the form of a price interval; and receiving user feedback to continuously optimize the system. The application realizes semantic-level understanding through field customization NLP technology, overcomes the lexical gap problem, provides multi-dimensional reliable price reference, supports batch automatic processing, and significantly improves the efficiency and accuracy of engineering material inquiry.

A processor-implemented method including generating input data based on raw text and a prompt for a predetermined domain, generating, through a large language model (LLM), event candidates corresponding to the raw text, verifying the event candidates based on structured knowledge as target events, normalizing the target events using information related to the predetermined domain, and extracting the normalized target events.

The application provides a software test case generation method and system based on domain knowledge constraints. A domain knowledge base covering domain basic knowledge and industry testing criteria is constructed; the original requirement text is normalized, the semantic understanding ability of the knowledge base and the large language model is combined, the requirement type is identified and the structured decomposition rule is dynamically matched, and a structured requirement model is constructed; the test focus features in the requirement are extracted, and the test item decomposition requirements are formulated; the large language model is driven to expand the scene and systematically disassemble the requirement, and the systematic test items are generated according to the test item decomposition requirements; the test cases are generated in multiple dimensions with the test items as the input, and the traceable relationship among the requirements, the test items and the test cases is established; the integrity check and consistency check are performed, and the test design result is output. The application enables the large language model to perform the test design task in a controlled and professional knowledge environment, and realizes the systematic generation from the natural language requirement to the executable test case.

This invention discloses a logic synthesis mapping method based on FPGA dual-output lookup tables. First, the target circuit netlist is preprocessed and structurally normalized, mapping logic nodes to single-output lookup tables of a unified format. Then, through primitive normalization, each lookup table unit is converted into native lookup table primitives consistent with the target FPGA technology library, forming a standardized lookup table netlist. An inverted index is constructed, and candidate lookup table pairs with shared inputs are generated using index relationships. Structural and functional constraints are sequentially performed on the candidate pairs to filter out mergeable lookup table pairs. After sorting them using a resource cost function, a merge mapping is performed sequentially, concatenating them into a six-input dual-output lookup table and generating corresponding initialization parameters. This method can reduce the number of lookup tables and input pin pressure while maintaining logical functional correctness, and can be integrated as an independent optimization step into various FPGA synthesis and place-and-route tools.

The application provides a frame parallel multimodal state level communication method, device and equipment for long video understanding and a storage medium, and relates to the technical field of artificial intelligence and computer vision. The method comprises the following steps: dividing an input video into a plurality of frame subsequences; based on a plurality of provider large language models, encoding the plurality of frame subsequences in parallel when sharing the same text context or task instruction to generate multimodal intermediate state representations; extracting visual key-value states therefrom, and performing distinguishing, screening and normalization processing to form a multimodal state key-value pair set; injecting the visual key-value states into a text condition key-value cache of a carrier large language model; based on the key-value cache after completing state injection, performing inference calculation by the carrier large language model to generate video understanding related output results. The application can guarantee communication efficiency while retaining and fusing key visual and timing information in the video more fully, and improve the overall inference performance in the long video understanding task.

This invention discloses an intelligent conversion method and system for natural language requirements to Selenium scripts based on a large language model, relating to the field of software engineering technology. The method includes: S1, requirement text normalization; S2, test element structuring; S3, enhanced context construction; S4, script template and location strategy generation; S5, initial script generation and execution; and S6, script adaptive optimization. This invention can accurately map unstructured natural language requirements to structured test elements and generate executable scripts. Throughout the process, it preserves sentence-by-sentence traceability and semantic alignment, allowing each automated test case to trace back to the original requirement. This significantly improves requirement-test consistency and reduces the risk of misjudgment and omission of test cases due to misunderstanding biases. By enhancing context representation, location fingerprints, and execution feedback loops, this solution enables automatically generated scripts to have runtime adaptive correction and self-healing capabilities.

A computer-implemented system and method for analyzing quality-event information through standardized representations, rule-based logic, and adaptive training. The system receives free-text quality-event descriptions and stores them in a historical database. A standardization module converts narratives into structured representations using domain terminology normalization and calibrated severity indicators. A rule-logic engine generates quality-rule constructs identifying primary, secondary, and systemic causes. A root-cause clustering module classifies events into related causal clusters. Based on identified clusters, the system generates training modules and control -measure templates for preventive or corrective actions. A simulation engine executes adaptive, branching scenarios presenting role-specific decision points and collecting performance metrics. A benchmarking engine computes comparative indicators across organizational units using anonymized data. An update module refines quality-rule constructs based on simulation performance and external system data. The system integrates with quality management, enterprise resource planning, and training platforms, enabling continuous organizational improvement through objective, data-driven analysis and dynamic, personalized training delivery.

The present application relates to the technical field of automatic report generation and data intelligence processing, and discloses a chapter-driven research report data acquisition and evidence pool construction method and system. It aims to solve the problems of insufficient chapter evidence support and low construction efficiency caused by the ambiguity of natural language requirement semantics and the blindness of multi-source retrieval. The technical scheme includes: receiving the original requirement text for standardized analysis and parameter completion, establishing a data requirement model to output a structured task object; decomposing the task into chapter-level retrieval targets and generating a retrieval plan; parallelly expanding the text and structured field collection tasks and performing hierarchical storage; calculating the sufficiency score based on the collection state and comparing it with the threshold to generate a dynamic supplementary collection decision; gathering the collection results to calculate the correlation score and construct candidate clusters, and outputting unified evidence entities after deduplication, normalization and conflict coordination; dividing the evidence pool based on the purpose of labeling and calculating the chapter adaptation degree to output the chapter adaptation mapping result.

One example method for improving the quality of responses generated by a virtual entity, such as a chatbot, in response to a user query includes, in response to a user query, retrieving content, and metadata associated with the content, from a table that includes cells, representing the content and metadata in a normalized data structure, determining, based on the normalized data structure, cell interdependencies of the table, and performing a content serialization process on the content to transform the content to a natural language structure.

This invention discloses a normalized log generation method based on the entropy increase principle, belonging to the field of database log auditing technology. The method includes: determining the entropy value of each field in the current log based on the entropy increase principle; determining whether a field is an important or unimportant field by using the relationship between the entropy value of the field and a set threshold; setting a related primary key for the current log, wherein the related primary key can represent the unique mapping relationship between all important fields and all unimportant fields in the current log; and converting the current log into a normalized log structure including important fields, related primary keys, and unimportant fields. This solution, using the entropy increase principle normalization method, can effectively handle original logs from multiple sources with different structures and has strong generalization ability. In addition, the final normalized log structure containing important and unimportant fields and related primary keys has a rigorous and unified format and concise rule definitions, and can be rapidly iterated and upgraded as the log structure changes.

Systems and methods provide fail-closed permit-before-actuate control for high-stakes effectors. A permit engine receives an operation request, canonicalizes operation context to a context digest, and, in an attested execution environment, evaluates a versioned policy pack, optionally with overlays, against the digest and commitments to evidence bundles under freshness thresholds. Upon compliance, a proof-of-policy-compliance receipt is produced, a minimal receipt core is derived, and a receipt commitment is generated. The commitment is anchored in an append-only continuity-verified log publishing signed heads with inclusion and consistency proofs subject to a maximum-merge-delay policy. A mint-after-verify process issues a scoped permit token only after verifying log proofs; the token is bound to the receipt commitment and a token scope and includes anti-replay and validity constraints. An actuation gate denies by default and enables actuation only after time-of-use validation of scope, revocation status, and proof freshness, and outputs precondition failure records with deterministic reason codes.

The application discloses a kind of FPGA excitation generation system and method based on semantic understanding and knowledge enhancement, system includes demand structured processing unit, knowledge retrieval and matching unit, test intention generation unit, code synthesis and check unit and feedback closed-loop control unit.Method includes: the joint analysis of multi-source test demand containing text and image is carried out, and standardization demand description text is generated;Through semantic vector coding and key word and semantic vector double-channel weight fusion retrieval strategy, the most similar historical test code and its function abstract are obtained from verification code knowledge base;Large language model is called to generate test function description text and initial FPGA test code in stages;Syntax check and format normalization processing are carried out to initial code;Through man-machine interaction, receive correction instruction and trigger iteration regeneration.The application effectively improves the analysis ability to multi-modal hardware test demand, enhances the reuse level to historical verification asset, improves the accuracy and engineering usability of generated test code, significantly reduces the cost of manual writing.

The application discloses a heterogeneous interface adaptation method based on a large language model and related products. The method comprises the following steps: obtaining interface field metadata according to acquired interface definitions and running evidence; obtaining context materials about an interface field for any interface field; obtaining prompt words according to the context materials; a large language model outputs semantic analysis results according to the prompt words; a mapping relationship is established according to the semantic analysis results and is added to a knowledge graph; a mapping relationship list of a source system and a target system is obtained according to the knowledge graph; a large language model is called, and structured mapping rules are obtained according to the mapping relationship list; and interface mapping specifications are obtained according to the structured mapping rules and are put on line through a static release process. In the embodiment of the application, the preparation and implementation cycle of interface connection is shortened, and the mapping efficiency is improved through semantic alignment and standardized output; the equivalence of different fields in a specific business scenario is understood by using a large language model, and the mapping accuracy is improved.

A method, system, and computer-readable medium are disclosed for processing and comparing documents in varying formats, incorporating advanced data extraction, normalization, and comparison techniques. The method includes receiving documents containing structured and unstructured data, processing them with optical character recognition and layout analysis to identify structural elements, and extracting key data fields using natural language processing, including named entity recognition and contextual analysis. Extracted data is normalized through schema mapping and synonym resolution, ensuring consistency in terminology and structure. The system employs fuzzy matching algorithms to identify discrepancies between documents and provides an interactive user interface with graphical visualizations, such as heatmaps and comparison charts, for reviewing and resolving differences. Machine learning techniques refine the system over time by integrating user feedback to improve data extraction, synonym resolution, and comparison accuracy.

A computer-implemented method of machine-learning. The method includes obtaining a training dataset of 3D models of real-world objects. The method further includes learning, based on the training dataset and on a patch-decomposition of the 3D models of the training dataset, a finite codebook of quantized vectors and a neural network. The neural network comprises a rotation-invariant encoder. The rotation-invariant encoder is configured for rotation-invariant encoding of a patch of a 3D model into a quantized latent vector of the codebook. The neural network further includes a decoder. The decoder is configured for decoding a sequence of quantized latent vectors of the codebook into a 3D model. The sequence corresponds to a patch-decomposition. This constitutes an improved solution for 3D model generation.

This application provides a method, apparatus, computer device, and storage medium for generating indictments, comprising: acquiring a text describing the facts of a case; inputting the text describing the facts of a case into a pre-trained language model to generate a corresponding indictment text; performing semantic similarity evaluation, structural length deviation evaluation, and key element coverage evaluation on the indictment text to obtain a first evaluation result, a second evaluation result, and a third evaluation result; calculating a comprehensive evaluation score based on the first evaluation result, the second evaluation result, and the third evaluation result; constructing a loss function based on the comprehensive evaluation score, and using the loss function to iteratively optimize the language model to obtain a target model; and inputting the target text describing the facts of the case to be processed into the target model to generate the target indictment text. Using the above method can improve the efficiency and quality of intelligent indictment generation, achieving standardized and high-quality automated indictment generation.

A system and a method are disclosed for identifying libraries used by an application based on the program code of the application. The system accesses a plurality of program codes of the application. For at least one program code, the system extracts raw type data from the program code and normalizes the raw type data to generate target normalized type data. The target normalized type data includes data type information of the program code. The system determines a set of candidate library types corresponding to the program code. The system accesses candidate raw type data associated with each candidate library type and determines a library type corresponding to the program code by comparing the program code of the application with the program code of each of the set of the candidate library types.

The application belongs to the field of software development application data processing, and particularly relates to a software development application data processing method based on an AI large model, which comprises the following steps: adopting a unified collection and text standardization processing method for first-line feedback and work order data, cleaning and labeling description information from each terminal; performing semantic analysis through an AI large model to extract elements such as equipment, process, function and problem type; adopting a hierarchical modeling and relationship extraction method for equipment, process and function information to construct an equipment-process-function relationship network; performing path traversal and weighted aggregation on text associated nodes to calculate the influence score of each software function; performing task template filling and historical case retrieval on high-score functions; and performing effect evaluation and parameter updating on the running results of the online development task; the application can uniformly process multi-source application data, quantify the influence of software functions and automatically generate development tasks, and realizes fine research and development decision-making.

The present application relates to the technical field of knowledge graph, in particular to a steel manufacturing knowledge graph construction method based on multi-source heterogeneous data, comprising: converting message data into standardized semantic triples through semantic mapping rules, and using physical and chemical component balance logic to complete missing attributes to form standardized entities; at the same time, using a semantic processor to extract the logic of unstructured report text, and performing cross-modal coupling with SEM image features to generate failure mechanism enhanced entities. Subsequently, collect full-process dynamic parameters and regularize them into equidistant time sequence parameters. Finally, taking the standardized entity as the index feature, taking the mechanism enhanced entity as the correlation constraint, using a topological feature mapping engine to perform semantic fusion, and combining a dynamic probability distribution correction operator to iterate parameter weights, a dynamic weight graph structure is constructed. The present application realizes efficient correlation and semantic organization of multi-modal data in the whole manufacturing process, greatly improving the retrieval accuracy and knowledge discovery ability of steel manufacturing data.