995 results about "Serialization" patented technology

The invention discloses a method and a system for supporting visual and AI bidirectional intercommunication editing views. The method comprises the following steps: establishing a bidirectional mapping rule between a rendering tree protocol and an intercommunication protocol; in response to an operation of a user on the visual editor, generating a first atomic operation instruction, applying the first atomic operation instruction to the current rendering tree and triggering local redrawing of a view; serializing the updated rendering tree into context segments in an intercommunication protocol format according to the mapping rule, and pushing the context segments to an AI model; responding to a natural language editing request initiated by a user to the AI model, receiving streaming data in an intercommunication protocol format returned by the user based on the request, analyzing the streaming data according to the mapping rule to obtain a second atomic operation instruction, applying the second atomic operation instruction to the rendering tree and triggering local redrawing of a view, and obtaining a second atomic operation instruction. The view is edited by the AI, so that real-time two-way intercommunication between visual editing and AI editing is realized, and the user experience is remarkably improved.

The invention belongs to the field of high-performance computing, and relates to an integer parallel computing method and device based on distributed storage and computer equipment, and the method comprises the steps of collecting real-time resource indexes, dynamically identifying fault nodes, triggering task migration, and performing data verification and hard disk fault detection. The weight value of each node is calculated, the nodes are arranged according to the descending order of the weight values, and the nodes with high load capacity are selected to distribute tasks; dynamically distributing a data generation task to a computing node, executing parallel computing, and performing distributed storage on a result; obtaining an operand, converting the operand into a first-order tensor form of a basic operand, serializing tensor data, and sending the serialized tensor data to a parallel computing layer; distributing a search task to a computing node, retrieving storage data in parallel, reading effective data from a storage layer, and combining search results into a partial sum; and summarizing and then outputting. The system has dynamic resource management and fault-tolerant capabilities, and can realize efficient task allocation and load balancing.

The invention provides a transformer substation operation and maintenance-based lock control terminal work log query method and system, and the method comprises the steps: collecting an original operation log stream generated by a transformer substation lock control terminal in a preset time period, serializing the original operation log stream to obtain a log data sequence, and then carrying out the context correlation analysis and recognition of an entity interaction relationship and behavior attribute description in a log text; a structured log metadata set is obtained after standardized integration; constructing a spatial topological index based on the physical layout and the electrical connection relationship of the substation equipment, mapping the equipment associated information in the spatial topological index to a corresponding node position of the spatial topological index, and generating a spatial enhanced log data set; receiving a natural language query request, generating a structured query expression, performing multi-dimensional retrieval matching on the spatial enhancement log data set, and outputting a preliminary query result set; and carrying out time sequence correlation analysis and operation influence evaluation on the log records to generate a query report. According to the invention, efficient and accurate query of the work log of the transformer substation lock control terminal can be realized.

The invention discloses a high-resolution remote sensing image semantic segmentation method based on multidirectional parallel selective scanning, and belongs to the technical field of high-resolution remote sensing image processing. According to the method, a multidirectional parallel selective scanning model is provided, direction perception modeling is carried out through 8-direction serialization scanning in combination with a state space model SSM, and the multidirectional long-distance dependence capture capability is enhanced while the linear calculation complexity is kept. A pyramid encoder-decoder structure is constructed, multi-level feature extraction is realized through a four-stage OSSBlock module, and local details and global semantics are dynamically fused in cooperation with SE attention jump connection of a decoder. A selective scanning mechanism is adopted to replace self-attention, and linear complexity calculation is realized through a state space parameter matrix; and designing a mixed loss function, and improving the small target segmentation precision in combination with the class balance of Dice Loss and the difficult sample mining capability of Focal Loss.

The invention relates to a hyperspectral image classification method and device based on a state space model. The hyperspectral image classification method based on the state space model comprises the following steps: sequentially carrying out feature extraction and serialization processing on hyperspectral image data to obtain a shallow feature projection vector; performing global-local feature extraction on the shallow feature projection vector by adopting a neural network based on a state space model to obtain a fused feature projection vector; and carrying out pixel-by-pixel classification and dimension rearrangement on the hyperspectral image data in sequence to generate a classification result of the hyperspectral image data. According to the hyperspectral image classification method based on the state space model, long-range dependence modeling is achieved through the neural network based on the state space model with linear complexity, the calculation complexity is effectively reduced, and through feature fusion and residual error connection, the classification accuracy of the hyperspectral image is improved. And the perception capability of the neural network on different scale space-spectrum structures in the hyperspectral image is effectively enhanced.

The invention belongs to the field of three-dimensional target detection, and particularly relates to a lightweight Mama three-dimensional target detection method based on columnar point cloud representation, which mainly consists of three parts: a spatial feature processing module based on Hilbert serialization and dynamic grouping, a Mama module based on a selective state space and a lightweight pyramid feature fusion module. The Hilbert serialization module keeps spatial locality through a serialization mode, and is helpful for capturing a structural relationship between adjacent points in the point cloud, so that the modeling effect of local feature integration and context information is improved; the Mama module adopts an input-dependent dynamic modeling mechanism, can provide higher modeling efficiency and stronger long-range dependent modeling capability when processing a large-scale volume column sequence, and is particularly suitable for a sparse point cloud scene; the lightweight pyramid fusion module enhances the perception capability of a target boundary through single-round up-down sampling and feature fusion, and improves the detection performance of a small target, thereby effectively balancing the efficiency and precision of 3D target detection.

The invention provides a shield muck three-dimensional point cloud segmentation and volume calculation method based on deep learning, and belongs to the technical field of deep learning and shield, and the method comprises the steps: S1, point cloud data collection, preprocessing and sample labeling; s2, constructing and training a deep learning segmentation model; s3, carrying out muck point cloud segmentation; and S4, calculating the volume of the muck. According to the method, efficient interaction of point cloud attention is realized by adopting four serialization modes, and the influence of point cloud disorder on calculation is avoided.

The invention provides a reward model training method and device, a strategy model training method and device, electronic equipment and a storage medium, and belongs to the technical field of artificial intelligence, and the method comprises the steps: obtaining a preference data pair which comprises a preferred response and a non-preferred response generated for the same prompt word, and each of the preferred response and the non-preferred response is composed of a plurality of text unit sequences; inputting each text unit sequence into a to-be-trained reward model to obtain a predicted reward value; and calculating the total training loss according to the predicted reward value, and updating the model parameters of the to-be-trained reward model. According to the method, the response text is subjected to serialized splitting, and the preference data composed of the preferred response and the non-preferred response is introduced for comparative learning, so that the target of model training is no longer to evaluate the absolute quality of a single response, but to identify a key text unit which causes one response to be superior to the other response; the fine-grained evaluation of the response content is realized, and the evaluation accuracy of the reward model and the identification capability of complex user preferences are effectively improved.

The invention discloses a PCB component layout method, device and equipment based on a multi-modal large model, and relates to the technical field of electronic design automation, the method comprises the following steps: carrying out quality screening on a plurality of PCB design files which have been laid out to obtain a qualified design file set; constructing a serialized training sample set based on the qualified design file set; each training sample sequence comprises a plurality of training samples arranged according to a layout logic sequence; each training sample comprises multi-modal data representing a current layout state and a real position coordinate of a next component to be laid out; based on the serialized training sample set, training the open-source multi-modal large model to obtain a PCB component layout model; and iteratively generating a layout file of the target PCB based on the netlist file of the target PCB and the PCB component layout model. The method is high in generalization capability and low in cost based on the PCB component layout model trained by the layout data, and can realize rapid layout of components of a new PCB.

The invention provides a dynamic traceability identifier generation and verification method based on multi-node environment data fusion, and relates to the technical field of information security and Internet of Things data credibility. The method comprises the following steps: firstly generating a seed identifier, then collecting environment data through each node to construct a multi-dimensional vector and carrying out weighted fusion, after serialization, generating a dynamic hash identifier by combining with an upper-level identifier, and meanwhile, introducing an offline trusted evolution, time drift tolerance and self-adaptive risk control mechanism; and finally, backtracking the traceability chain through recursion verification, and guaranteeing privacy in combination with zero-knowledge proof. The method solves the problems that a static traceability code is easy to copy and the data credibility is low, achieves the dynamic anti-counterfeiting and tamper-proofing of the traceability identifier, enables the digital identifier to be strongly associated with physical circulation, improves the credibility of the traceability process, can automatically find abnormal circulation and give an alarm, and is suitable for a commodity traceability scene.

The invention discloses an intelligent twin regulation and control method and system based on material process optimization. A multi-node regulation and control mechanism is set along a time axis, and equipment parameters, process parameters and environment parameters are collected and serialized into a multi-dimensional input sequence; synchronously acquiring spectral characteristics and performance indexes of the material, and constructing a spectrum-performance output sequence; establishing a many-to-many parameter simulation model based on a Transform architecture, and analyzing a dependency relationship among parameters by using a multi-head attention mechanism; a spectrum-efficiency data set is extracted from literatures in combination with a semantic model, prediction analysis and optimization parameter setting are carried out on real-time / simulation parameters, finally process adjustment is guided through prediction results and parameters, and a closed-loop feedback iterative optimization simulation model is formed. According to the method, material process dynamic modeling, precise prediction and regulation and control analysis are realized, the process optimization efficiency and the material performance prediction precision are remarkably improved, and the research and development trial and error cost is reduced.

The invention discloses a PDF contract file identification method and system, a medium and a program product, and relates to the technical field of information identification, and the method comprises the steps: carrying out the image analysis of an obtained PDF contract file through employing a deep learning model, generating an editable text, and synchronously identifying the page layout of the document, outputting structured data including page numbers, text paragraphs and coordinate information; performing multi-dimensional feature matching on the text paragraphs according to an adaptive semantic analysis algorithm, a preset contract template library and a dynamic keyword library, and positioning contract core element information; performing entity relationship verification on the content of the contract core element information, correcting an extraction error through dependency syntactic analysis and semantic role labeling, and establishing a contract element data set containing a confidence coefficient weight; and performing serialized packaging on the verified contract element data set to generate a contract element list conforming to electronic signature authentication. According to the invention, the processing efficiency and recognition precision of the PDF contract file are improved.

A computer-implemented method and apparatus for managing physical and digital products' customization using container format are disclosed. In some embodiments, the method comprises receiving a container file with serialized objects, deserializing these objects to form a hierarchical object structure, and interacting with a database to obtain real-time configuration data. Custom attributes of the products are defined by binding attribute / value pairs to the hierarchical structure. A visualization of the customized product is generated based on these pairs, and manufacturing instructions are determined considering the attribute / value pairs and manufacturing constraints. The apparatus comprises a memory and a processor to execute instructions for deserializing the container file, retrieving configuration data, organizing objects into a hierarchical tree, negotiating product customizations, and outputting manufacturing data. This system enables efficient and accurate production of customized products by managing data, rendering graphics, and incorporating manufacturing constraints.

The invention relates to the technical field of artificial intelligence and automatic testing, and particularly discloses an AI-based automatic testing method and device. The method comprises the following steps: capturing and recording complete execution historical information in a process of executing a test task by an AI agent, and serializing and storing the complete execution historical information as reusable cache data; before the test is executed, the validity of the cache is judged through a multi-dimensional verification mechanism including data integrity, timeliness, functionality and the like; when the cache verification is passed, a result is quickly obtained based on the cache data playback test process; when cache verification is not passed or execution fails, triggering an AI agent to perform new reasoning and steps to generate and update cache data; and meanwhile, a caching strategy is dynamically adjusted according to test case changes and resource conditions. The device comprises a test executor, an AI agent manager, an intelligent cache manager and a dynamic cache management module, and all the modules work cooperatively to improve the test efficiency, reduce the calling cost of a large language model and enhance the stability of a test result. The method is suitable for Web / APP automatic testing, cross-platform testing, CI / CD process testing and other scenes, the execution efficiency can be remarkably improved, the cost is reduced, and resource utilization is optimized.

The invention discloses a distributed job scheduling method and device, a storage medium and computer equipment, relates to the technical field of distributed task scheduling, is suitable for the field of financial and medical services, and mainly aims at solving the problem that an existing distributed job scheduling system lacks task dependence management and dynamic scheduling capability. Comprising the following steps: performing task modeling processing on each to-be-executed task by adopting a binary tree structure to obtain a task binary tree corresponding to each to-be-executed task; performing serialization processing on the task binary tree by adopting a job arrangement engine to obtain a corresponding task sequence; issuing the task sequence to an event flow bus; after the event flow bus is started, tasks to be executed are allocated to different distributed actuators for execution based on the central control scheduler; and the distributed executor returns an execution result to the event flow bus in real time, so that the event flow bus triggers the downstream task after the upstream task is completed until the task sequence is completely executed.

The invention discloses a multi-dimensional data asynchronous calculation method and system based on a dynamic dependency graph, and relates to the field of data processing. The method comprises the following steps: constructing a metadata, logic and instance three-layer separation storage model; analyzing the reference relationship to construct a directed acyclic graph, and generating a calculation priority; monitoring data change, and executing asynchronous serialization calculation through a message queue based on priority; the associated document is automatically updated based on anchor mapping. According to the method, the calculation deadlock and the performance bottleneck of large-scale data in the Web environment are solved, logic decoupling and dynamic expansion are realized, the final consistency of the data is guaranteed, and the high-concurrency throughput and the stability are remarkably improved.

A computer-implemented system for routing artificial intelligence (AI) queries. The system utilizes a zero-copy data pipeline, which processes prompts in memory-mapped buffers to eliminate at least one memory copy operation, thereby reducing latency relative to conventional serialization pipelines. The system continuously verifies AI provider compliance by injecting synthetic prompts containing invisible, Ed25519-signed Unicode watermarks. Algorithmic bias is detected by generating counterfactual “digital twin” prompts and applying Fisher exact statistical testing.Routing decisions for multi-tier autonomous systems are governed by safety-level requirements (ASIL-D, ASIL-B, QM) and may be constrained by external routing directives received via a meta-identifier. A hash-chained manifest, cryptographically signed using Ed25519 and consumed by downstream gateways, is generated for each routing decision, with its Merkle root asynchronously anchored to a blockchain to create a tamper-evident audit trail for regulatory compliance.

The invention belongs to the technical field of data processing, and particularly relates to a fingerprint generation method of a virtual evaluation model based on Hash operation, which comprises the following steps of: firstly, constructing a dependency tree by taking a main model file specified by a user as a root node, and initializing an access record table; establishing a static keyword protection area containing a core parameter rule base, scanning a dynamic file operated by a user, and comparing parameter definitions to generate an early warning; reading the binary data of the current node file, and generating a character string through SHA-256 after standardization; recursively processing the sub-nodes to obtain a node tree hash value, sorting the sub-nodes, splicing the sub-nodes according to a fixed format, and generating a character string through SHA-256 operation; then, a current node standardized path is spliced and generated through SHA-256 operation; and finally, by taking the root node as a starting point and the DFS as a sequential serialization tree structure, executing SHA-256 operation after adding a global collision salt value, and outputting a global unique fingerprint of the model. The problems of poor stability, low efficiency and high generation risk in a model fingerprint generation technology in the prior art can be solved.

The invention belongs to the technical field of vehicle software testing, and particularly relates to a test case generation method and system based on big language model and rule collaboration, and the method comprises the steps: analyzing a demand document, and mapping the demand document to a fault ontology model to generate a structured test demand; generating a plurality of groups of fault verification conditions based on equivalence class division and a boundary value analysis method by utilizing a large language model; a multi-dimensional mixed retrieval strategy is adopted to retrieve a reference case from a historical case library; in combination with the verification condition, the reference case and the standardized information base, preposed environment settings, serialized operation instructions and expected responses of the test case are generated in stages; and finally, the case quality is ensured through cooperative verification of a large language model and a rule engine. The problems that a traditional method depends on manpower, the coverage rate is low, errors are prone to occurring, and historical knowledge is difficult to reuse are effectively solved, and the automation degree, accuracy and integrity of BMS test case generation are remarkably improved.

The invention discloses an industrial internet data management method and system, and relates to the technical field of industrial internet. By deploying a data acquisition module at an edge node, real-time acquisition and time serialization processing of data are realized, the problems of real-time performance and distribution of data processing are solved, through a distributed streaming processing framework and a parallelization analysis technology, the data processing speed is improved, and it is ensured that equipment state abnormity can be found and responded in time; by constructing a data-driven decision support model and performing trend prediction and optimization suggestion generation on high-value data, the problems of data value evaluation and resource allocation are solved, the resource utilization efficiency is improved, and finally, the suggestion priority is adjusted by adopting a weighted analysis method in combination with historical data of resource optimization, so that the resource optimization efficiency is improved. And a structured data file for subsequent decision-making reference is generated, so that the intelligent decision-making capability is remarkably enhanced.

The invention relates to the field of machine learning, particularly discloses an artificial intelligence-based power system fault analysis and diagnosis method, and effectively solves the problem of information loss caused by neglecting a key waveform form in a transient signal in the prior art through a local feature extraction and serialization module. An original signal is converted into a local feature sequence with more characterization significance. Aiming at the averaging bottleneck of an existing model in an information aggregation stage, a traditional feature compression method is abandoned, and a sequence information aggregation and decision-making mechanism is provided. According to the mechanism, a context sensing sequence is regarded as a probability event, and modeling is carried out on the sequence from three orthogonal dimensions of a content center, time sequence dispersion and distribution uncertainty by calculating feature expectation, time sequence variance and information entropy of the context sensing sequence. The method can deeply insight and quantify the essential difference of different events in the time sequence dynamic evolution mode, thereby fundamentally solving the problem of misjudgment caused by feature confusion.

The invention discloses an intelligent extraction method and equipment for structural data of a scanned copy and a medium, and relates to the cross technical field of computer vision and intelligent document processing. The method comprises the following steps: acquiring a to-be-processed scanned copy image, and performing joint feature extraction on the scanned copy image based on a preset multi-modal neural network model to generate intermediate feature representation; constructing an extensible structured index template library containing field rules, value domain constraints and cross-field verification logic; based on an attention mechanism, performing intelligent semantic matching and mapping on the intermediate feature representation and a target field in an extensible structured index template library to output preliminary structured data; based on a verification rule preset in an extensible structured index template library, executing automatic logic verification and conflict resolution on the preliminary structured data to generate standard structured data conforming to business specifications; and carrying out serialization output on the standard structured data according to a predetermined format.

The invention relates to the field of traffic storage optimization and traffic data compression, in particular to a multithreading-combined lossless compression method for fine-grained traffic, which comprises the following steps of: acquiring a traffic data file, extracting a header triple based on the traffic data file, generating an identifier according to the header triple, and transmitting the identifier to a server; aggregating the traffic data files with the same identifier to obtain data streams, and generating a sorted data stream size table according to the data volume of the data streams; creating threads according to system computing power, distributing the data flow to the thread with the minimum thread load according to the data flow size table and the thread load condition, and outputting a data flow distribution scheme; and performing fine-grained characterization and serialization on each data stream to obtain an integer sequence and a byte sequence, selecting a compression processing method according to redundancy characteristics of the integer sequence and the byte sequence, obtaining a redundancy-eliminated sequence, and writing the redundancy-eliminated sequence into a compressed file. The invention aims to provide high compression efficiency to reduce the storage cost while ensuring the data precision.

The invention discloses a real-time target detection model and method fusing multi-scale feature enhancement and dynamic label distribution. The real-time target detection model sequentially comprises a data preprocessing unit, a backbone network, a semantic detail injection feature fusion module, a ShareSePhead detection head, a dynamic label distribution unit, a training module and a post-processing unit. After the backbone network is subjected to convolution processing, three feature maps with different scales are output; the semantic detail injection feature fusion module converts the multi-layer features into serialized features and forms a unified fusion feature tensor; the detection head uniformly extracts features among the scales through a shared convolutional layer, and outputs a classification prediction vector and bounding box regression result; the training module is used for training and updating parameters; and the post-processing unit is used for executing filtering operation. According to the method, the reasoning speed is remarkably improved, and high-frame-rate real-time detection is realized; network computing resource allocation is optimized, and a better cost-effectiveness ratio is achieved.

This specification relates to privacy-preserving model training on tabular data. In some aspects, a method includes receiving, by one or more computing devices, tabular data; serializing the tabular data into a natural language string in a natural language format; combining the natural language string and a prompt as an input to a pretrained large language model (LLM) to generate a predicted result, wherein a set of learned vectors are added into the pretrained LLM for fine-tuning the pre-trained LLM; fine-tuning the pretrained LLM using a differential privacy stochastic gradient descent (SGD) process, wherein fine-tuning the pretrained LLM comprises: determining values of the learned vectors that minimize a difference between the predicted result and the ground truth; receiving a request including test tabular data for a predication task; and generating, in response to the request for the prediction task, a prediction result for the test tabular data using the fine-tuned LLM.

A high-speed transmitter system using a common mode controlled serialization stage embedded in an output stage is disclosed. In some embodiments, the transmitter includes a serialization circuit that is configured to convert parallel data into serial data with one or more serialization stages; a logic circuit that is configured to connect each input of a last stage of the one or more serialization stages of the serialization circuit, via a respective logic function, to a respective dedicated output stage of an output circuit; and the output circuit that is configured to implement output stages to generate a signal based on the received input using NMOS transistors. In addition, the logic circuit is configured to be a desired voltage value (e.g., the output circuit's common mode) such that a switching point of the NMOS gm stage can be decreased and the gm stage can be activated within a reduced time.

The invention discloses a small sample unified granularity relation extraction method based on a large language model, which comprises the following steps of: firstly, giving a specific task description as a part of an input context of the large language model; thirdly, giving an analogous example to the large language model as context demonstration; in order to better prompt the position information of the entity of the large language model in the context, performing entity enhancement on the context input into the large language model; and finally, a mode for serializing the relation triad is defined, and thinking chain reasoning information is fused in the mode, so that a large language model can be helped to perform relation extraction by utilizing thinking chain prompts. According to the method, context learning, thinking chain and entity enhancement technologies are introduced for unified granularity relation extraction tasks including a sentence level, a document level and a cross-document level, the powerful reasoning ability of a large language model is fully played, and the effectiveness of the model in the unified granularity relation extraction task, especially in a small sample scene, is improved.

The invention provides a business process model instance construction method and device based on a data structure, and the method comprises the steps: carrying out the assembly according to a JSON structure and a business process drawn by a user, and generating a business process JSON data structure; performing data serialization on the business process JSON data structure to generate a database process draft basic information extension table; in response to a release process instruction input by a user, performing data deserialization according to the database process draft basic information extension table, generating business process series information and updating a business process series table; in response to a process starting instruction input by a user, determining a business process series information instance from the business process series table; according to the business process model architecture, the business process series information instances are spliced, and the business process model instance is generated, so that expansibility, maintainability and flexibility can be improved, front-end load work is reduced, complexity and error risks are reduced, and response speed is increased.

The invention provides a structured memory data processing method and system oriented to long time sequence interaction, and is applied to the technical field of natural language processing and artificial intelligence memory modeling. The method comprises the following steps: acquiring a real-time interaction data stream, partitioning the real-time interaction data stream into dialogue data blocks, loading a hierarchical attribute mode and a previous time step attribute tree instance, inputting serialized texts of the hierarchical attribute mode and the previous time step attribute tree instance into a generative model, generating a writing, rewriting, deleting or null operation instruction aiming at a leaf node path, analyzing, updating, generating a current attribute tree instance, and storing the current attribute tree instance; by means of the scheme, lossy compression and structured evolution of the infinite long dialogue stream can be achieved, memory forgetting is relieved on the premise that a context window is not expanded, and long-time-sequence information retrieval precision and storage efficiency are improved.

The invention discloses an end-to-end lightweight road crack sensing method, system and device and a medium, belongs to the technical field of computer vision and deep learning, and aims to solve the technical problem of how to quickly and accurately detect road cracks and improve the road crack detection efficiency. According to the technical scheme, the method comprises the following steps: image standardization processing: carrying out standardization preprocessing on a collected road image to obtain a road image after standardization preprocessing; multi-scale dynamic labeling: labeling the road image after standardization preprocessing by adopting a multi-scale dynamic labeling mechanism to construct a training label; light-weight road crack perception: based on an improved DETR model, introducing a cross attention mechanism into a Transform encoder to improve the perception ability of crack local features, and adopting a serialization interaction mechanism in the Transform encoder to realize crack integrity reasoning; and transfer learning incremental training: improving the generalization performance of the model through a transfer learning strategy combining public data set pre-training and own data set incremental training.